Tissue expansion is a powerful technique that allows reconstruction of defects with tissue of similar tone, color, and texture, and results in minimal donor-site morbidity.1 , 2 Although it was initially described as a modality for ear reconstruction in the pediatric population, tissue expansion has evolved toward a wide variety of applications, allowing achievement of aesthetic and functional endpoints that were not previously obtainable with other methods of reconstruction.3 Current indications for the use of tissue expansion include treatment of burn scars, giant congenital nevi, aplasia cutis congenita, microtia, meningomyelocele, hemangioma, midfacial cleft, Romberg disease, Poland syndrome, scrotal reconstruction, and vaginal agenesis, among others.4–8
Soft-tissue expansion may be used to increase the surface area of full-thickness skin grafts, local and regional flaps, or free flaps before transfer. Evidence has shown that expanded tissue has similar durability and contracture compared to nonexpanded tissue when harvested for a full-thickness skin graft or flap.9 Expansion should ideally be started early (before school age) to avoid peer pressure that may develop later in childhood because of the changes in the patient’s appearance with expander filling. Scalp expansion should be delayed until at least 9 months of age to minimize molding of the pliable cranium by the tissue expander.
Despite the increasing use of tissue expansion in a variety of clinical applications, complications are common, with rates as high as 40 percent in children and infants quoted in earlier reports.4 However, these reports did not deter the acceptance of tissue expansion as a major reconstructive modality. In our experience, we have found that although complications are inherent in the process of tissue expansion, they do not preclude meeting the goals of reconstruction. We present the 12-year experience of the senior author (A.K.G.) in the use of tissue expansion in the pediatric population. We describe the major and minor complications encountered, and identify risk factors and make recommendations on means of avoiding complications. We also propose treatment and salvage strategies in the event that complications should arise in the process of tissue expansion.
PATIENTS AND METHODS
Institutional review board approval was obtained to conduct a retrospective chart review of the senior author’s (A.K.G.) patients over a 12-year period. All patients who underwent tissue expansion performed by the author during the study period were included in the analysis. All patients were operated on and followed at a single institution. Patient charts were reviewed to determine patient age at the time of initial expander placement, diagnosis, number of expanders placed, expander location, expander type, capacity of the expander, and volume instilled relative to capacity. Indications for expansion included giant congenital nevi (most common presenting complaint in this series), burn reconstruction, hemangioma, cutis aplasia, keloids, dog bite injuries, Pfeiffer syndrome, nevus sebaceous, and bladder exstrophy. Complications encountered in the process of tissue expansion were categorized as major or minor. Major complications were defined as loss of expander before completion of the reconstruction caused by expander rupture or exposure. Minor complications were defined as those in which the expander was not explanted and the reconstructive goals were achieved. The results were analyzed using binomial and multifactorial logistic regressions in IBM SPSS (IBM Corp., Armonk, N.Y.).
Expanders were placed under general anesthesia, with perioperative antibiotic administration to cover skin flora. Both rigid and flexible-backed expanders were placed. Seamless (flexible-backed) expanders were placed under flaps located over bony prominences such as the calvaria or rib cage. Rigid expanders were placed in areas where control of expansion vector was more difficult, such as the abdominal wall. Integrated internal ports were used in a limited fashion in this series. Double-stage expanders (i.e., expanders in which the base unit expands first and then the upper unit expands) were used earlier in the series, but placement ceased when an inordinate number of exposures and infections resulted.
The considerations used in flap design included, among other factors, the region of the body expanded, the likelihood of serial expansion, and the stage of the reconstruction. Incisions were placed radially (perpendicular) relative to the direction of expansion. If an incision was placed near the potential defect, the pocket was dissected as far from the lesion as possible. The expander pocket was dissected wider than the base of the expander in a blunt fashion to preserve the longitudinal blood supply. In anticipation of contraction following expansion, the expander was planned to create a flap 30 to 50 percent longer than necessary when maximally filled. Rather than choosing an expander size arbitrarily, we chose the expander with the largest surface area possible to expand the tissue available for transposition.
The expander was placed on top of the deep fascia unless underlying muscle was integrated into the flap. Scalp expanders were placed in a subgaleal pocket. Because of the propensity for complications in the lower extremities previously described in the literature,10 , 11 few expanders were placed inferior to the buttocks. Back-cuts were used in a very limited number of patients in the lower extremities. The expanded flaps were secured in a layered fashion with interrupted 3-0 Monocryl (Ethicon, Inc., Somerville, N.J.) sutures in the deep fascia and interrupted 4-0 or 5-0 nylon skin sutures. No drains were placed.
If expander placement was performed for reconstruction of a giant congenital nevus, the incision was placed least 2 cm inside the nevus border to remain remote from the expander. The incisions were made parallel to the edge of the nevus for better access to dissect the expander pocket and to avoid tension. The length of the incision was made shorter than the length of the expander to leave undissected tissues that would prevent subsequent transposition of the expander. The authors feel that it is more difficult to control the expander pocket and subsequent expander transposition if the incision is placed radially within the nevus. The expander pocket was dissected beneath normal skin, extending to, but not beyond, the junction of normal skin and nevus except in the situation of serial expansion for giant congenital nevus.
In the case of serial expansion, placement of subsequent expanders occurred 3 months after flap advancement. Earlier in the series, we did attempt immediate expander re-placement after flap advancement, but we found that the expander pocket could not be reliably controlled in this situation and had a high risk of implant migration. Tissue expansion commenced 3 to 4 weeks after expander placement, with expanders filled up to two times the recommended capacity. Because many patients resided far away from the hospital, expansion was performed routinely by a patient’s family member, primary care physician, or a physician extender after adequate education, with regular review by the senior author. The t test was used for all statistical analyses.
The retrospective chart review showed that 94 pediatric patients underwent expansion with a total of 282 expanders placed during the 12-year study period. The mean age of patients at the time of initial expander placement was 39 months (range, 7 to 163 months). A total of 65 complications were encountered, of which 37 were major and 28 were minor. Major complications were defined as those requiring expander removal, such as infection, rupture, and exposure. Minor complications were defined as those that did not necessitate expander removal, but did delay or affect the course of the reconstruction in some way. The complication rate per patient was 41.5 percent (39 complications per 94 patients), which was higher than the 23.0 percent complication rate per expander (65 complications per 282 expanders), because of many patients undergoing serial expansion.
The types of complications encountered are listed in Table 1. Rupture was most commonly encountered in the scalp, accounting for nine of 20 complications (45 percent) in this region. Other complication data are shown by anatomical site of expansion in Table 2 and by expander fill and chronicity in Table 3. According to our protocol, expander fill never exceeded twice the recommended volume. As shown in Table 3, underfilled expanders had the highest (60.7 percent) complication rate. Contrary to intuition, overfilling of expanders was instead associated with only a 14.6 percent complication rate, not much higher than the 9.8 percent rate encountered when expanders were filled to the recommended volume. To illustrate this, exposure occurred in eight of 56 underfilled expanders (14 percent) compared to one of 185 overfilled expanders (0.5 percent). We believe, however, that the higher complication rate in underfilled expanders is not necessarily inherent in the fill of the expander, but perhaps is caused by the natural course of complications such as infection, which declare themselves before the surgeon has had the chance to fill the expander to or beyond capacity.
Double-stage expanders were used early in the series but were subsequently abandoned, as they were associated with a higher complication rate compared with single-stage expanders. Of six double-stage units placed, four had either migrated or become exposed. This complication rate of 67 percent was much higher than the 21 percent complication rate for single-stage expanders (58 complications in 276 single-stage expanders placed).
Patients underwent either primary (n = 195) or serial (n = 87) tissue expansion. There was a statistically significant difference (p = 0.012) between the complication rate encountered per expander for primary (19.0 percent) and serial expansion (32.2 percent). However, there was no statistically significant difference (p = 0.178) between the 33.3 percent complication rate encountered with immediate placement of expanders after advancement of a prior expanded flap or the 30.6 percent complication rate encountered when subsequent expander placement was delayed until after the flap had healed. The complication rates during successive rounds of serial expansion are illustrated in Figure 1. Five of the 94 patients required five or more rounds of expansion. Importantly, there was a marked rise in the complication rate after the second round of serial expansion, with expanders placed in the sixth round or later suffering a 100 percent complication rate. Based on this experience, it is our recommendation that serial expansion should be limited to four rounds if possible, and should not proceed to or beyond a sixth round of expansion.
Statistical analysis of the results using a binomial regression showed a strong correlation between the presence of any complication and the expander location (p < 0.005) or the stage of serial expansion (p < 0.005). Multivariate regression analysis showed no statistical significance between the type of complication and the expander location (p = 0.410) or the stage of serial expansion (p = 0.298).
Despite the complications encountered in the process of tissue expansion, the vast majority of patients went on to achieve their reconstructive goals. Figure 2 illustrates a patient undergoing scalp expansion in preparation for excision of a congenital nevus. Her course was complicated by expander rupture, but after removal of the expander and flap advancement, the defect site could be closed and the reconstruction was successfully completed. Expander exchange with subsequent resumption of tissue expansion can be used to manage cases of expander folding, as in Figure 3, which shows a 5-year-old boy who underwent expander placement in the abdomen. The expander had folded and created pressure points in the skin and could no longer be inflated. Exchange of the expander allowed continuation of the expansion protocol and provided adequate skin for both donor- and recipient-site closure in one stage.
Where expander exchange is not possible, the reconstructive goals may still be met using tissue available to resurface critical areas of the defect and harvesting additional tissue as required. This principle is illustrated in Figures 4 and 5. Figure 4 illustrates a patient who was undergoing an expanded full-thickness skin graft from the abdomen to resurface the dorsum of the hand. Because of exposure of the abdominal tissue expander, further expansion was stopped and a full-thickness skin graft was harvested from the expanded lower abdominal skin. An additional split-thickness skin graft was sufficient to resurface the rest of the defect involving the wrist and forearm. Figure 5 illustrates an alternative method of salvage following expander exposure when the expander cannot be exchanged. A 6-year-old girl who underwent tissue expansion in the scalp to reconstruct the hairline following severe burns to the face and scalp is shown. Expansion was complicated by expander exposure. The expander was removed, and flap advancement and rearrangement were used to compensate for the defect caused by exposure of the expander, with a satisfactory cosmetic result.
Our experience with tissue expansion in the pediatric population compares favorably with recent studies, which report complication rates ranging from 13 to 30 percent, with complication rates as high as 65 percent reported in the literature.4–6 , 11–14 This study presents one of the largest series to date with regard to tissue expansion in the pediatric population, and is unique in that it reflects the experience of a single surgeon (A.K.G.).
Overall complications by anatomical region in our series appeared to be greatest in the torso, followed by the head and neck. Minimal complications were encountered in the lower extremities, although this is also because placement was avoided in the lower extremities unless absolutely necessary. Our experience, supported by other series in the literature, indicates that expansion below the waist is exceedingly problematic in the pediatric population.5 , 13 We also found that tissue expansion of the scalp was consistently associated with a high incidence of complications. It is reasonable to hypothesize that expanders in this location are prone to disruption because of vulnerability during falls or other trauma, and because of their location on the fixed calvarial surface. The gross irregularities that are present on the calvaria as a result of remodeling may also impinge on the expander, resulting in an additional stress and increasing the likelihood of disruption and rupture.
The type of expander placed appears to influence the complication rate of the reconstruction. Double-stage expanders were associated with a significantly higher complication rate compared with single-stage devices, with migration and expander exposure being prevalent. One double-stage expander “flipped” while being filled, such that expansion had to be terminated. Placement of double-stage expanders in the present series was therefore limited based on the observed complications with these expanders. Likewise, expanders with incorporated ports were seldom used because of excessive thinning of the flap overlying the port site, particularly in areas when the skin was tight, such as in the scalp. In addition, there were concerns that expander puncture may occur during port access, the risk of which would be increased in an uncooperative child. Interestingly, we found that overfilling an expander had no adverse effects on reconstruction, and thus this is an option for achieving sufficient tissue surface area for reconstruction.
We found that serial expansion is associated with an increased risk of expander-related complications. This is consistent with a previous study by Friedman et al.,11 who reported that a history of two or more prior expansions resulted in a major complication in 40 percent of patients. Vergnes et al.15 reported a decreased net gain of 50 percent for each new expansion in preparation for excision of congenital nevi in the trunk. In contrast, Hudson et al.16 advocated the safety of serial tissue expansion, and reported major complication rates of 20, 18, and 0 percent for the second, third, and fourth expansions, respectively; however, the limited sample size of 11 expanders for the third expansion and three expanders for the fourth expansion indicates that one must be cautious in drawing conclusions from these data. Our experience suggests that serial expansion should not proceed beyond the third round of expansion if possible, and should definitely not proceed beyond the fifth round of expansion.
Based on our experience in this study and from the published literature, we can make several recommendations in an effort to develop guidelines in salvage of complications, which thus far appear to be unavoidable in the process of tissue expansion. In the event of expander exposure, a logical treatment plan would involve removal of the expander and replacement if possible. Where expander exchange is not possible because of skin loss, flaps can be designed and advanced to distribute redundancy into the region of exposure. Risk factors for exposure include rigid backed expanders in the scalp, and serial expansion. Thus, these should be avoided where possible.
In the event of expander rupture, if the expander has been filled to less than the recommended fill volume, we have found that expander exchange with immediate refill serves to achieve the reconstructive goals. If the expander has been filled to greater than the recommended fill volume, the expander should be removed and flaps designed and advanced to compensate for loss of skin and subcutaneous tissue. A risk factor for rupture appears to be placement of expanders in the scalp; in addition, some patients by nature of their activities (e.g., sports, tumbling) have an increased risk of expander damage. Thus, tissue expansion should be used with caution in these patients.
In the event of infection, the expander should be removed with immediate advancement of flaps so as not to lose gains made from expansion to date. Where serial expansion is contemplated, this should be delayed until the infection has completely resolved. We have found that risk factors for infection include prior infection and use of serial expansion. To minimize risk of infection with serial tissue expansion, aseptic technique and use of perioperative antibiotics are of utmost importance. Although oral antibiosis may be attempted if the surgeon feels that the infectious process is a cellulitis and is confined to the skin, we believe that any infectious process involving the expander per se should be treated aggressively, as described above.
With expander migration, replacement of the expander is indicated only if further expansion would be ineffective. However, migration of the expander may predispose to other complications such as exposure, and therefore patients should be followed more closely. In the torso, where expander migration was more common, one option is to place bolsters around the periphery of the expander for 2 to 3 weeks to allow the capsule to secure the expander location before initiating expansion. Alternatively, expander inflation can be delayed up to 3 months to allow the capsule to mature. These techniques are particularly useful in cases of serial expansion in the torso where expanders are being placed in previously expanded beds.
In the event of port malfunction, use of a remote port precludes the need for expander removal and requires only port exchange. This may be encountered in situations where the port can flip, becoming inaccessible. Moreover, as mentioned above, a port access error is less likely to damage the vulnerable expander when a remote port is used. This is especially important if non–plastic surgery–trained providers or family members are administering the expansion fills at home when the patient is away from the primary surgeon’s practice. Therefore, we recommend the use of expanders with remote ports for all pediatric patients.
Tissue expansion in the pediatric population is an effective reconstructive modality, despite potential complications. Patients and, more importantly, their families must be made aware that up to one-third of patients will have a complication necessitating additional surgery or modification of the initial reconstructive plan. Many families perform expansion at home, and therefore they should be made familiar with common complications such as expander exposure, rupture, and infection to eliminate delay in subsequent treatment. Patients and families who are cooperative and compliant are most likely to have a better outcome, and anticipatory guidance, psychological support, and education are thus vital components of the treatment process. Although the majority of reported complications of tissue expansion serve to deter or delay the final reconstruction, few complications prevent the achievement of reconstructive goals. Proper patient selection, thorough preoperative planning, meticulous technique, and the capacity to modify the reconstructive plan for each patient based on their clinical response and complications are all critical factors for achieving an optimal outcome.
The authors thank Timothy Santoro, M.D., for assistance in reviewing charts of the study patients.
Parents or guardians provided written consent for the use of patients’ images.
1. Argenta LC, Marks MW, Pasyk KA. Advances in tissue expansion. Clin Plast Surg. 1985;12:159–171.
2. Vander Kolk CA, McCann JJ, Knight KR, O’Brien BM. Some further characteristics of expanded tissue. Clin Plast Surg. 1987;14:447–453.
3. LoGiudice J, Gosain AK. Pediatric tissue expansion: Indications and complications. J Craniofac Surg. 2003;14:866–872.
4. Elias DL, Baird WL, Zubowicz VN. Applications and complications of tissue expansion in pediatric patients. J Pediatr Surg. 1991;26:15–21.
5. 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;38:358–364.
6. Iconomou TG, Michelow BJ, Zuker RM. Tissue expansion in the pediatric patient. Ann Plast Surg. 1993;31:134–140.
7. Paletta C, Campbell E, Shehadi SI. Tissue expanders in children. J Pediatr Surg. 1991;26:22–25.
8. Rivera R, LoGiudice J, Gosain AK. Tissue expansion in pediatric patients. Clin Plast Surg. 2005;32:35–44, viii.
9. Bauer BS, Vicari FA, Richard ME, Schwed R. Expanded full-thickness skin grafts in children: Case selection, planning, and management. Plast Reconstr Surg. 1993;92:59–69.
10. Bauer BS, Vicari FA, Richard ME. The role of tissue expansion in pediatric plastic surgery. Clin Plast Surg. 1990;17:101–112.
11. Friedman RM, Ingram AE Jr, Rohrich RJ, et al. Risk factors for complications in pediatric tissue expansion. Plast Reconstr Surg. 1996;98:1242–1246.
12. 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;57:93–97.
13. Pisarski GP, Mertens D, Warden GD, Neale HW. Tissue expander complications in the pediatric burn patient. Plast Reconstr Surg. 1998;102:1008–1012.
14. Youm T, Margiotta M, Kasabian A, Karp N. Complications of tissue expansion in a public hospital. Ann Plast Surg. 1999;42:396–401; discussion 401402.
15. Vergnes P, Taieb A, Maleville J, Larrègue M, Bondonny JM. Repeated skin expansion for excision of congenital giant nevi in infancy and childhood. Plast Reconstr Surg. 1993;91:450–455.
16. Hudson DA, Lazarus D, Silfen R. The use of serial tissue expansion in pediatric plastic surgery. Ann Plast Surg. 2000;45:589–593; discussion 593594.
Coding perspective provided by Dr. Raymond Janevicius is intended to provide coding guidance.
- 11960 Insertion of tissue expander(s) for other than breast, including subsequent expansion
- 11971 Removal of tissue expander(s) without insertion of prosthesis
- The insertion of a tissue expander is reported with code 11960.
- Code 11960 is used for the placement of tissue expanders in all non-breast anatomic sites. The code does not distinguish between sizes of expanders.
- The use of the term “expander(s)” is sometimes misinterpreted by payers to include all expanders placed during one operative session. If one pocket is created, and more than one expander is placed in that pocket, then code 11960 is reported once. This is the intent of the use of the plural term “expander(s).”
- If, however, more than one pocket is created and separate expanders are placed in each pocket, then code 11960 is used for each expander placed, as creation of additional pockets requires additional surgical work.
- If four scalp tissue expanders are placed in four separate pockets, the procedures are reported:
- All office visits for tissue expansion during the 90-day global period are included in code 11960. If tissue expansion is performed beyond the 90-day global period, appropriate office visit codes (9921X) are reported.
- The removal of a tissue expander is reported with code 11971.
- If the tissue expander is removed within the 90-day global period of code 11960, then modifier 58 is appended: 11971-58.
- Creation of one pocket for expander(s).
- Insertion of expander(s) into one pocket.
- Intraoperative tissue expansion.
- Wound closure.
- Tissue expansions during the 90-day global period.
- Code 11971 does not include:
- Lesion excisions.
- Reconstructive procedures (e.g., flaps, complex repairs).
Disclosure: Dr. Janevicius (firstname.lastname@example.org) is the President of JCC, a firm specializing in coding consulting services for surgeons, government agencies, attorneys, and other entities.©2018American Society of Plastic Surgeons