- Email: [email protected]
Minimal invasive reconstruction of posttraumatic hemi facial atrophy by 3-D computer-assisted lipofilling
Journal of Plastic, Reconstructive & Aesthetic Surgery (2007) 60, 1138e1144
Minimal invasive reconstruction of posttraumatic hemi facial atrophy by 3-D computer-assisted lipofilling* C. Hoehnke*, M. Eder, N.A. Papadopulos, A. Zimmermann, G. Brockmann, E. Biemer, L. Kovacs Department of Plastic and Reconstructive Surgery, Klinikum rechts der Isar, Technical University Munich, Ismaninger Str. 22, 81675 Munich, Germany Received 5 December 2005; accepted 30 January 2007
KEYWORDS Hemi facial atrophy; Lipofilling; 3-D scanning; Quality control
Summary Facial hemi atrophy is seen after trauma, Parry-Romberg syndrome and on other rare occasions. Since the aesthetic deficit is very obvious and irritating, facial reconstruction is often requested by these patients. In most cases the only option for sufficient reconstruction is free flap reconstruction, which represents the standard treatment. Recently in plastic surgery, various new techniques have been developed with the potential for multiple applications. Lipofilling has been presented primarily for the correction of cosmetic lesions or the reconstruction of minor soft tissue defects, but even reconstruction of larger soft tissue deficits is possible. The concept of using 3-D technology in facial reconstruction has multiple advantages. Primarily, the ideal final aesthetic outcome can be simulated by virtual reconstruction. Mathematic calculations deliver exact numbers in volume deficits, enabling precise planning of soft tissue substitution especially in lipofilling, ideally avoiding unnecessary corrections. Since autologous soft tissue reconstruction represents a dynamic process with periods of swelling as well as atrophy, quality control is required for achieving optimal results. Use of 3-D scanning has the advantage of reliable visualisation in soft tissue reconstruction without the limitations of harmful side effects. We present the history of a female suffering from the posttraumatic consequences of head injuries related to a car accident and the successful correction of her hemi facial atrophy by autologous lipofilling. Technical details and the potential of 3-D scanning in plastic surgery are discussed. ª 2007 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.
* Part of this work was presented at the Annual Meeting of the German Association of Plastic Surgeons (VDPC) in Munich, Germany, held between September 28th and October 1st, 2005. * Corresponding author. Tel.: þ49 89 4140 2171; fax: þ49 89 4140 4869. E-mail address: [email protected] (C. Hoehnke).
Facial atrophy may occur uni- or bilaterally. Bilateral facial atrophy is found in various situations such as trauma,1 HIV infection2,3 and Barraquer-Simons syndrome.4 Facial hemi atrophy is seen after trauma,5 Parry-Romberg syndrome6 and on other rare occasions.7,8
1748-6815/$ - see front matter ª 2007 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.bjps.2007.01.068
Minimal reconstruction of posttraumatic hemi facial atrophy The aesthetic deficit is very obvious and irritating particularly in facial hemi atrophy. For this reason facial reconstruction is often requested by patients with this condition. In most cases, the only option for sufficient reconstruction is free flap reconstruction, which represents the standard treatment. Recently in plastic surgery, various new techniques have been developed with the potential for multiple applications. Lipofilling has been primarily presented for the correction of cosmetic lesions or the reconstruction of minor soft tissue defects. Three-dimensional assessment of facial deformities by the new imaging technologies provides useful information in planning and evaluating facial reconstruction. Based on positive results with these techniques1 we have taken advantage of the combination of both lipofilling and 3-D soft tissue surface scanning in the reconstruction of a rare case of posttraumatic hemi facial atrophy. Technical details and the potential of 3-D computerised tomographic scanning in plastic surgery are discussed.
Case report A 45-year-old lady presented with a hemi facial atrophy on the right side. She reported that she had had a car accident abroad 15 years ago, resulting in severe head injuries including various facial fractures. Consequently she had undergone multiple operations, finally ending up with a hemi facial atrophy on the right side. The patient primarily wanted a cosmetic improvement of the tissue defect, since she suffered psychologically from her facial deformity. The correction of the nose was explicitly declined. Clinical and radiological examination of the face showed a severe asymmetry of the face with soft tissue deficits in the area of the zygoma and the cheek. Moreover, the nose had a deviation to the right resulting in an oblique midline axis. The function of the mimic muscles, however, was not impaired. In addition, former fractures of the nose, upper orbital edge, zygoma and the chin area of the right mandible were revealed. At the right orbit and chin the bony defect healing had resulted in the formation of a step. Neurologic examination and neurography as well as myography of the facial nerve revealed no irregularities. Otolaryngologic studies were normal except for the deviation of the nose to the right including a subluxation of the nasal septum. The result of ophthalmologic examinations was normal in relation to the age of the patient. The patient was informed of the options in hemi facial reconstruction, especially the standard treatment of free flap transfer, but also of the more recent concept of soft tissue reconstruction by autologous fat transplantation as a multi-step procedure. Since the patient refused invasive surgery (including rhinoplasty) she decided to undergo multi-step lipofilling. Based on solid personal experience with 3-D imaging in facial reconstruction and promising reports in the literature, we used a Minolta Vivid 910 (Minolta Co., Ltd, Osaka, Japan) laser scanner in setting up the treatment concept. In order to create a virtual scaffold of the ideal right side of the face, the uninjured left side of the face was mirrored. By calculating the difference in volume between the original left side and the computer animated right side
1139 of the face it was possible to express the local soft tissue deficit in numbers (Fig. 1 and Table 1). Moreover, a small but clearly visible defect was defined in the medial upper lid region. Since the original chin was very prominent, the defect on the right chin area was even more obvious. The total volume deficit was calculated to be 120 cc. Resorption was assumed at 20%. As a result, the total volume required for optimal augmentation was calculated to be 146 cc. In order to achieve optimal results and a high percentage of viable fat cells, the amount of fat cells grafted was limited. The first operation was performed in October 2003. After harvesting the fat in the abdomen, 26 cc were injected into the zygoma and 28 cc into the cheek area. Additionally 8 cc were placed in the right chin area. This made a total of 62 cc. After an initial swelling and moderate haematoma, the postoperative course was uncomplicated. The effect of the first operation was clearly visible 3 months later and could be visualised as well as quantified by 3-D scanning. Since the patient had gained weight, the soft tissue deficit analysis was 49 cc. At the second operation, 12 cc were injected into the zygoma, 22 cc into the cheek and again 8 cc in the chin area. Moreover, 5 cc were injected into the medial upper lid zone. During the postoperative course no complications were observed. After 12 months, the midface was almost symmetric. Only the chin and upper lid area could be further improved. Comparison studies by 3-D scanning between the results at 3 and 9 months after the second operation showed only minimal differences. 3-D scanning just before the third operation demonstrated an additional augmentation of 33 cc. Finally, in January 2005 an additional 6 cc were injected into the medial upper lid, 10 cc into the zygoma, 15 cc into the cheek and 6 cc into the chin area. At the beginning of this operation, two fat cylinders had been harvested from the cheek area by a special set normally used for liver biopsies. The histology of this fat sample showed that it was living fat tissue. The aesthetic outcome three months after the third operation was very pleasing. In all four zones of autologous fat transplantation a dramatic improvement in comparison to the primary situation has been visibly achieved. This subjective view is supported by the latest results of 3-D scanning, where a significant augmentation in all four zones was calculated with a total of 118 cc. Only in the chin area did additional improvement seem to be realistic.
Material and methods Three-dimensional computerised scanning Conventional documentation of changes in the facial outlook of the patient was performed by 2-D photography. Additionally, 3-D scans were taken pre- and postoperatively by a 3-D-linear-laser-scanner (Minolta Vivid 910 3D Digitizer, Konica-Minolta Co., Ltd, Osaka, Japan). Documentation of the face by a 3-D-linear-laser-scanner requires three single shots in different positions: one frontal 0 picture, and two additional pictures showing the right and left facial region in 30 angle to the sagittal level according to a standard protocol.9,10
1140
C. Hoehnke et al.
Figure 1 (a) Based on the simulated ideal right side of the face (Fig. 2 a) differences with the preoperative condition are calculated and shown in various colours according to a millimetre scale. (b) These horizontal and lateral cuts demonstrate the differences between the calculated ideal and the final outcome 6 months after the last operation. Since there are only minimal differences, an almost optimal reconstruction has been achieved.
The single shots were composed as 3-D pictures and analysed by the Raindrop Geomagic Studio 7 and Qualify 7 software (Raindrop Geomagic, Inc. NC, USA). In order to analyse facial symmetry of the patient, the scan of the intact left side of the face was mirrored to the injured right side. Consequently, the healthy right side of the face represented the point of reference in further analyses for reconstruction of facial symmetry by operative procedures. At the first preoperative level, a 3-D comparison of the mirrored healthy left side of the face with the injured right Table 1 Calculated soft tissue deficit versus injected volume of autologous fat transplantation in cc OP I (Oct 2003) Med. upper lid Zygoma Cheek Chin Sum Total
27/26 29/28 8/8 64/62
OP II (Jan 2004)
OP III (Jan 2005)
4/5
5/6
13/12 24/22 8/8 49/47
2/10 4/15 5/6 16/37
129/146
side was performed; differences at the marked spots were calculated in mm by Raindrop Geomagic software. Moreover, volume differences were calculated by superposed pictures in cm3. Additional sagittal cuts demonstrated any differences in two dimensions. After every operation, the preoperative 3-D scans were compared with the postoperative results. Basically, the left side of the face represented the ideal form of the right side, and was used more or less as a scaffold. In the end, the final operative result of the right side of the face was compared with the mirrored left side.
Harvesting and injecting the fat The fat was harvested in all operations by liposuction from the abdomen through an umbilical access using the Coleman technique. Fat injection was performed using a personally modified Coleman technique.
Results The soft tissue deficit before the first operation was calculated to be 64 cc. With 62 cc, almost the exact amount
Minimal reconstruction of posttraumatic hemi facial atrophy of fat was injected. Since the total injected fat volume was relatively high, an overcorrection was not performed, in order to achieve an optimal survival rate of the fat, especially avoiding oil cysts. After the first operation, the patient gained 8.5 kg weight. Consequently all calculations before the next operation were based on the ‘new’ data of the
1141 contralateral side of the face. Three months after the first operation, analysis demonstrated a soft tissue deficit of 49 cc. In the second operation, the total amount of injected fat was 49 cc. For personal reasons, the patient wanted to wait one year for the next operation. At that time, the preoperative analysis showed a soft tissue deficit of 16 cc.
Figure 2 These photographs demonstrate along the horizontal axis the frontal/lateral view and the calculated soft tissue deficit. In the vertical axis, pre- and postoperative conditions are shown.
1142 Since this operation was intended to be the final one (using 37 cc), an overcorrection was performed on this occasion. The visual comparison of the pre- and postoperative pictures (Fig. 2) clearly shows an improvement in terms of symmetry. Of particular note is the fact that the virtual mirror image is almost identical to the final postoperative view (Fig. 3).
Discussion Generations of plastic surgeons have been searching for an ideal method to replace or augment soft tissue. Ideally, a substitute would look and feel like the original tissue intended to be replaced. Moreover, no, or minimal, donor site morbidity would be involved. Last but not least, the procedure should be easy to perform. A special indication for soft tissue reconstruction is facial atrophy. Currently, in most of the cases, the only option for sufficient hemi facial reconstruction is free flap reconstruction.11e13 Theoretical alternatives are represented by the implantation of various materials such as gold, silver, wires, rubber, paraffin, silk, gutta-percha and ivory.14e16 In the middle of the last century, the use of free or solid silicon started a new era.17,18 In the last two decades, interest has focused more and more on liquids that can be easily injected.19 Historically, autologous fat transplantation was successfully performed at the end of the 19th century. The first scientific study on the survival of free fat transplantation was published in 1950 by Peer, and reported a survival rate of 50%.20 Vascularisation of fat droplets occurs after 4 days, before which the cells survive by diffusion. Several authors have published reports of permanent survival of fat grafts.21,22 Others have shown that traumatic handling of fat results in partial or complete resorption.23e25 Despite these reports, many plastic surgeons and other physicians still favour the transplantation of non-vital fat. In 1999, Coleman reported positive experiences with autologous facial fat transplantation in over 10 000 patients with follow ups of up to 8 years.26 Lipofilling has been presented primarily for the correction of cosmetic lesions or the reconstruction of minor soft tissue defects. Recently, however, the indication has been enhanced for larger soft tissue deficits or the
C. Hoehnke et al. improvement of facial proportions. Since aesthetic operations represent non-essential surgery it is very important to meet patients’ expectations in terms of the final result. For this reason, some surgeons take more than 200 photographs from various angles to get a feeling for the individual amount of fat required to augment local soft tissue deficits.26 Based on this problem we were looking for a method which would deliver reliable and reproducible data about the amount of soft tissue deficit, as well as provide a 3-D simulation of the optimal outcome. For many years, industrial settings have used a number of technologies for recording and visualising 3-D objects. Recently, these technical principles have been transferred into the medical field. The first applications have shown promising results.27 Three-dimensional visualisation of body parts enables a better understanding of the anatomy. In soft tissue defects, comparison studies with the normal counterpart or the ideal anatomy can deliver digital data on the defect or the tissue component to be reconstructed. With a background of successful experiences with 3-D planning in skull reconstruction and major facial defects27 we have taken advantage of the combination of both fat transplantation and 3-D assessment of the surface geometry by 3-D laser scanning in the reconstruction of this rare case of posttraumatic hemi facial atrophy. In order to obtain a model of the ideal right side of the face, 3-D scanning was used to mirror the non-injured left side resulting in a symmetrical face. The only problem lay in the nasal area, due to the deviation of the nose to the right. Based on the calculation of the local volume differences between the ‘ideal’ left side and the injured right side of the face, a topographic map of the soft tissue deficit was produced (Fig. 1). The total volume of the soft tissue deficit was primarily calculated to be 64 cc. Since the patient gained weight after the first treatment, the calculated soft tissue deficit before the second operation was higher than expected. The soft tissue deficit calculated for the third operation was about one-third of the injected fat volume in the second operation. This rate correlates with the rate of resorption. Consequently, in the last operation an over-correction of 21 cc was performed. The postoperative development of so-called ‘oil
Figure 3 Assuming that the mirrored left side of the face represents the ideal of the right side, (Fig. 3b) represents such a simulation in comparison to the preoperative situation in (Fig. 3a) and (Fig. 3c) where the final outcome is shown.
Minimal reconstruction of posttraumatic hemi facial atrophy cysts’ is a common problem in lipofilling. Since in our case this complication was seen particularly after locally injecting high volume fat deposits, we postulated that insufficient vascularisation and injection of traumatised or nonliving fat cells were the basic problems. Consequently, all injections were performed using a personal minimal traumatic technique involving three principles: (1) prior to all fat injections, local anaesthetics with adrenaline were administered, thus reducing local haematoma and postoperative pain which may result in high blood pressure and consecutive bleeding. (2) Fat was only deposited in micro channels created immediately beforehand, thus minimising trauma to the individual fat cells. (3) Three-dimensional fat deposition was achieved by a multi-layer ‘woven fat fabric’. Swelling returned to normal after about 8 weeks. Resorption was calculated at about 30%. Manual palpation of the reconstructed right side of the face in July 2005, 6 months after the last operation, showed only moderate differences compared to the left side of the face. Histological examinations of cheek biopsies taken in the third operation, 15 months after the first operation, confirmed the presence of vital fat tissue with only minimal fibrosis. With respect to facial symmetry, the latest photographs demonstrate significant improvement, especially in the mid face. Objective quality control was achieved by 3-D scanning technology. The final results of a precise mathematic comparison in surface geometry of facial soft tissue as well as volumetric differences support the subjective visual impression of facial symmetry improvement. Further correction could be achieved by an additional silicon implant in the chin, a septo-rhinoplasty and bilateral otoclisis. Since the patient was absolutely satisfied with the operative result, she declined any further intervention. The concept of using 3-D technology in facial reconstruction has multiple advantages. Primarily, the ideal final aesthetic outcome can be simulated by virtual reconstruction. Mathematic calculations deliver exact numbers in terms of volume deficits, enabling precise planning of soft tissue substitution especially in lipofilling, which ideally avoids any unnecessary corrections. Since autologous soft tissue reconstruction represents a dynamic process with periods of swelling as well as atrophy, quality control is required to achieve optimal results. Use of 3-D scanning has the advantage of reliable visualisation in soft tissue reconstruction without the limitations of harmful side effects. There are also new potential pathways in facial reconstruction. Although only one patient history is presented here as a case report, the results are absolutely convincing. The presented concept is even appropriate in HIV-related atrophy, rare syndromes and also in senile atrophy as well as wrinkles. In terms of treatment costs, this noninvasive concept of facial reconstruction has the significant advantage of reducing donor site morbidity and hospitalisation or absence from work. In conclusion, precise planning of minimally invasive hemi facial reconstruction and consecutive quality control can be achieved by using 3-D scanning technology in combination with complex lipofilling.
1143
References 1. Watzinger F, Wanschitz F, Wagner A, et al. Computer-aided navigation in secondary reconstruction of post-traumatic deformities of the zygoma. J Craniomaxillofac Surg 1997;25: 198e202. 2. Talmor M, Hoffmann LA, La Trenta GS. Facial atrophy in HIVrelated fat redistribution syndrome: anatomic evaluation and surgical reconstruction. Ann Plast Surg 2002;49:11e7. 3. Hodgkinson DJ. Facial atrophy in HIV-related fat redistribution: anatomic evaluation and surgical reconstruction. Ann Plast Surg 2003;50:328. 4. Gueganton C, Chavoin JP, Boutault F, et al. Treatment of facial lesions in Perry-Romberg and Barraquer-Simons syndromes: report of 12 cases. Ann Chir Plast Esthet 2000;45:436e51. 5. Kawano Y, Araki E, Arakawa K, et al. A case of progressive hemifacial atrophy with Pourfour de Petit syndrome which was successfully treated by stellate ganglion block. Rinsho Shinkeigaku 1999;39:731e4. 6. Mazzeo N, Fisher JG, Mayer MH, et al. Progressive hemifacial atrophy (Parry-Romberg-Syndrome). Case report. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1995;79:30e5. 7. Gambichler T, Kreuter A, Hoffmann K, et al. Bilateral linear scleroderma ‘‘en coup de sabre’’ associated with facial atrophy and neurological complications. BMC Dermatol 2001; 1:9. 8. Roddi R, Riggio E, Gilbert PM, et al. Progressive hemifacial atrophy in a patient with lupus erythematosus. Plast Reconstr Surg 1994;93:1067e72. 9. Kovacs L, Zimmermann A, Brockmann G. Three-dimensional recording of the human face with a 3D laser scanner. J Plast Reconstr Aesthet Surg 2006;59:1193e202. 10. Kovacs L, Zimmermann A, Wawrzyn H, et al. Computer aided surgical reconstruction after complex facial burn injuries opportunities and limitations. Burns 2005;31:85e91. 11. Siebert JW, Anson G, Longaker MT. Microsurgical correction of facial asymmetry in 60 consecutive cases. Plast Reconstr Surg 1996;97:354e63. 12. Longaker MT, Siebert JW. Microsurgical free-flap correction of severe hemifacial atrophy. Plast Reconstr Surg 1995;96: 800e9. 13. Lutz BS, Toussaint S, Wei FC. Bilateral lipoatrophy secondary to connective tissue panniculitis treated with two microsurgically transplanted latissimus dorsi muscles. Ann Plast Surg 1998;40: 302e7. 14. Beck J. Implantation method: Plastic surgery of the nose and ear. In: Loeb H, editor. Operative Surgery of the Nose, Throat, and Ear. St Louis: CV Mosby; 1917. p. 3. 15. Miller C. The limitations and use of paraffin in cosmetic surgery. Wis Med Recorder 1908;11:277. 16. Rogers B. A chronologic history of cosmetic surgery. Bull N Y Acad Med 1971;47:265. 17. Brown JB, Fryer MP, Randall P. Silicones in plastic surgery. Plast Reconstr Surg 1953;12:374e6. 18. Rees TD, Ashley FL. Treatment of facial atrophy with liquid silicone. Am J Surg 1966;111:531e5. 19. Hinderer UT, Escalona J. Dermal and subdermal tissue filling with fetal connective tissue and cartilage, collagen and silicone. Aesthet Plast Surg 1990;14:239e48. 20. Peer LA. Loss of weight and volume in human fat grafts. Plast Reconstr Surg 1950;5:217. 21. Moscona R, Ullmann Y, Har-Shai Y, et al. Free-fat injections for the correction of hemifacial atrophy. Plast Reconstr Surg 1989; 84:501e7. 22. Niechajev I, Sevcuk O. Long term results of fat transplantation: clinical and histological studies. Plast Reconstr Surg 1994;94: 496e506.
1144 23. Ellenbogen R. Invited comment on autologous fat injection. Ann Plast Surg 1990;24:297. 24. Ersek R. Transplantation of purified autologous fat: A 3-year followup is disappointing. Plast Reconstr Surg 1991;87:219e27. 25. Illouz YG. Fat injection: a four-year clinical trial. In: Hetter GP, editor. Lipoplasty: The Theory and Practice of Blunt Suction Lipectomy. Boston: Little Brown; 1990. p. 239e46.
C. Hoehnke et al. 26. Coleman SR. Fettunterspritzung (Lipostructure). In: LemperleA¨sthetische Chirurgie 1999;VI-2.1 [ecomed, 3. Erg Lfg]. 27. Kovacs L, Zimmermann A, Brockmann G, et al. Three-dimensional surface scanning of the facial region with the Minolta Vivid-910 3D Scanner: a survey of examining conditions, precision and accuracy using a dummy model. IEEE Trans Med Imaging 2006;25:742e54.
Minimal invasive reconstruction of posttraumatic hemi facial atrophy by 3-D computer-assisted lipofilling* C. Hoehnke*, M. Eder, N.A. Papadopulos, A. Zimmermann, G. Brockmann, E. Biemer, L. Kovacs Department of Plastic and Reconstructive Surgery, Klinikum rechts der Isar, Technical University Munich, Ismaninger Str. 22, 81675 Munich, Germany Received 5 December 2005; accepted 30 January 2007
KEYWORDS Hemi facial atrophy; Lipofilling; 3-D scanning; Quality control
Summary Facial hemi atrophy is seen after trauma, Parry-Romberg syndrome and on other rare occasions. Since the aesthetic deficit is very obvious and irritating, facial reconstruction is often requested by these patients. In most cases the only option for sufficient reconstruction is free flap reconstruction, which represents the standard treatment. Recently in plastic surgery, various new techniques have been developed with the potential for multiple applications. Lipofilling has been presented primarily for the correction of cosmetic lesions or the reconstruction of minor soft tissue defects, but even reconstruction of larger soft tissue deficits is possible. The concept of using 3-D technology in facial reconstruction has multiple advantages. Primarily, the ideal final aesthetic outcome can be simulated by virtual reconstruction. Mathematic calculations deliver exact numbers in volume deficits, enabling precise planning of soft tissue substitution especially in lipofilling, ideally avoiding unnecessary corrections. Since autologous soft tissue reconstruction represents a dynamic process with periods of swelling as well as atrophy, quality control is required for achieving optimal results. Use of 3-D scanning has the advantage of reliable visualisation in soft tissue reconstruction without the limitations of harmful side effects. We present the history of a female suffering from the posttraumatic consequences of head injuries related to a car accident and the successful correction of her hemi facial atrophy by autologous lipofilling. Technical details and the potential of 3-D scanning in plastic surgery are discussed. ª 2007 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.
* Part of this work was presented at the Annual Meeting of the German Association of Plastic Surgeons (VDPC) in Munich, Germany, held between September 28th and October 1st, 2005. * Corresponding author. Tel.: þ49 89 4140 2171; fax: þ49 89 4140 4869. E-mail address: [email protected] (C. Hoehnke).
Facial atrophy may occur uni- or bilaterally. Bilateral facial atrophy is found in various situations such as trauma,1 HIV infection2,3 and Barraquer-Simons syndrome.4 Facial hemi atrophy is seen after trauma,5 Parry-Romberg syndrome6 and on other rare occasions.7,8
1748-6815/$ - see front matter ª 2007 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.bjps.2007.01.068
Minimal reconstruction of posttraumatic hemi facial atrophy The aesthetic deficit is very obvious and irritating particularly in facial hemi atrophy. For this reason facial reconstruction is often requested by patients with this condition. In most cases, the only option for sufficient reconstruction is free flap reconstruction, which represents the standard treatment. Recently in plastic surgery, various new techniques have been developed with the potential for multiple applications. Lipofilling has been primarily presented for the correction of cosmetic lesions or the reconstruction of minor soft tissue defects. Three-dimensional assessment of facial deformities by the new imaging technologies provides useful information in planning and evaluating facial reconstruction. Based on positive results with these techniques1 we have taken advantage of the combination of both lipofilling and 3-D soft tissue surface scanning in the reconstruction of a rare case of posttraumatic hemi facial atrophy. Technical details and the potential of 3-D computerised tomographic scanning in plastic surgery are discussed.
Case report A 45-year-old lady presented with a hemi facial atrophy on the right side. She reported that she had had a car accident abroad 15 years ago, resulting in severe head injuries including various facial fractures. Consequently she had undergone multiple operations, finally ending up with a hemi facial atrophy on the right side. The patient primarily wanted a cosmetic improvement of the tissue defect, since she suffered psychologically from her facial deformity. The correction of the nose was explicitly declined. Clinical and radiological examination of the face showed a severe asymmetry of the face with soft tissue deficits in the area of the zygoma and the cheek. Moreover, the nose had a deviation to the right resulting in an oblique midline axis. The function of the mimic muscles, however, was not impaired. In addition, former fractures of the nose, upper orbital edge, zygoma and the chin area of the right mandible were revealed. At the right orbit and chin the bony defect healing had resulted in the formation of a step. Neurologic examination and neurography as well as myography of the facial nerve revealed no irregularities. Otolaryngologic studies were normal except for the deviation of the nose to the right including a subluxation of the nasal septum. The result of ophthalmologic examinations was normal in relation to the age of the patient. The patient was informed of the options in hemi facial reconstruction, especially the standard treatment of free flap transfer, but also of the more recent concept of soft tissue reconstruction by autologous fat transplantation as a multi-step procedure. Since the patient refused invasive surgery (including rhinoplasty) she decided to undergo multi-step lipofilling. Based on solid personal experience with 3-D imaging in facial reconstruction and promising reports in the literature, we used a Minolta Vivid 910 (Minolta Co., Ltd, Osaka, Japan) laser scanner in setting up the treatment concept. In order to create a virtual scaffold of the ideal right side of the face, the uninjured left side of the face was mirrored. By calculating the difference in volume between the original left side and the computer animated right side
1139 of the face it was possible to express the local soft tissue deficit in numbers (Fig. 1 and Table 1). Moreover, a small but clearly visible defect was defined in the medial upper lid region. Since the original chin was very prominent, the defect on the right chin area was even more obvious. The total volume deficit was calculated to be 120 cc. Resorption was assumed at 20%. As a result, the total volume required for optimal augmentation was calculated to be 146 cc. In order to achieve optimal results and a high percentage of viable fat cells, the amount of fat cells grafted was limited. The first operation was performed in October 2003. After harvesting the fat in the abdomen, 26 cc were injected into the zygoma and 28 cc into the cheek area. Additionally 8 cc were placed in the right chin area. This made a total of 62 cc. After an initial swelling and moderate haematoma, the postoperative course was uncomplicated. The effect of the first operation was clearly visible 3 months later and could be visualised as well as quantified by 3-D scanning. Since the patient had gained weight, the soft tissue deficit analysis was 49 cc. At the second operation, 12 cc were injected into the zygoma, 22 cc into the cheek and again 8 cc in the chin area. Moreover, 5 cc were injected into the medial upper lid zone. During the postoperative course no complications were observed. After 12 months, the midface was almost symmetric. Only the chin and upper lid area could be further improved. Comparison studies by 3-D scanning between the results at 3 and 9 months after the second operation showed only minimal differences. 3-D scanning just before the third operation demonstrated an additional augmentation of 33 cc. Finally, in January 2005 an additional 6 cc were injected into the medial upper lid, 10 cc into the zygoma, 15 cc into the cheek and 6 cc into the chin area. At the beginning of this operation, two fat cylinders had been harvested from the cheek area by a special set normally used for liver biopsies. The histology of this fat sample showed that it was living fat tissue. The aesthetic outcome three months after the third operation was very pleasing. In all four zones of autologous fat transplantation a dramatic improvement in comparison to the primary situation has been visibly achieved. This subjective view is supported by the latest results of 3-D scanning, where a significant augmentation in all four zones was calculated with a total of 118 cc. Only in the chin area did additional improvement seem to be realistic.
Material and methods Three-dimensional computerised scanning Conventional documentation of changes in the facial outlook of the patient was performed by 2-D photography. Additionally, 3-D scans were taken pre- and postoperatively by a 3-D-linear-laser-scanner (Minolta Vivid 910 3D Digitizer, Konica-Minolta Co., Ltd, Osaka, Japan). Documentation of the face by a 3-D-linear-laser-scanner requires three single shots in different positions: one frontal 0 picture, and two additional pictures showing the right and left facial region in 30 angle to the sagittal level according to a standard protocol.9,10
1140
C. Hoehnke et al.
Figure 1 (a) Based on the simulated ideal right side of the face (Fig. 2 a) differences with the preoperative condition are calculated and shown in various colours according to a millimetre scale. (b) These horizontal and lateral cuts demonstrate the differences between the calculated ideal and the final outcome 6 months after the last operation. Since there are only minimal differences, an almost optimal reconstruction has been achieved.
The single shots were composed as 3-D pictures and analysed by the Raindrop Geomagic Studio 7 and Qualify 7 software (Raindrop Geomagic, Inc. NC, USA). In order to analyse facial symmetry of the patient, the scan of the intact left side of the face was mirrored to the injured right side. Consequently, the healthy right side of the face represented the point of reference in further analyses for reconstruction of facial symmetry by operative procedures. At the first preoperative level, a 3-D comparison of the mirrored healthy left side of the face with the injured right Table 1 Calculated soft tissue deficit versus injected volume of autologous fat transplantation in cc OP I (Oct 2003) Med. upper lid Zygoma Cheek Chin Sum Total
27/26 29/28 8/8 64/62
OP II (Jan 2004)
OP III (Jan 2005)
4/5
5/6
13/12 24/22 8/8 49/47
2/10 4/15 5/6 16/37
129/146
side was performed; differences at the marked spots were calculated in mm by Raindrop Geomagic software. Moreover, volume differences were calculated by superposed pictures in cm3. Additional sagittal cuts demonstrated any differences in two dimensions. After every operation, the preoperative 3-D scans were compared with the postoperative results. Basically, the left side of the face represented the ideal form of the right side, and was used more or less as a scaffold. In the end, the final operative result of the right side of the face was compared with the mirrored left side.
Harvesting and injecting the fat The fat was harvested in all operations by liposuction from the abdomen through an umbilical access using the Coleman technique. Fat injection was performed using a personally modified Coleman technique.
Results The soft tissue deficit before the first operation was calculated to be 64 cc. With 62 cc, almost the exact amount
Minimal reconstruction of posttraumatic hemi facial atrophy of fat was injected. Since the total injected fat volume was relatively high, an overcorrection was not performed, in order to achieve an optimal survival rate of the fat, especially avoiding oil cysts. After the first operation, the patient gained 8.5 kg weight. Consequently all calculations before the next operation were based on the ‘new’ data of the
1141 contralateral side of the face. Three months after the first operation, analysis demonstrated a soft tissue deficit of 49 cc. In the second operation, the total amount of injected fat was 49 cc. For personal reasons, the patient wanted to wait one year for the next operation. At that time, the preoperative analysis showed a soft tissue deficit of 16 cc.
Figure 2 These photographs demonstrate along the horizontal axis the frontal/lateral view and the calculated soft tissue deficit. In the vertical axis, pre- and postoperative conditions are shown.
1142 Since this operation was intended to be the final one (using 37 cc), an overcorrection was performed on this occasion. The visual comparison of the pre- and postoperative pictures (Fig. 2) clearly shows an improvement in terms of symmetry. Of particular note is the fact that the virtual mirror image is almost identical to the final postoperative view (Fig. 3).
Discussion Generations of plastic surgeons have been searching for an ideal method to replace or augment soft tissue. Ideally, a substitute would look and feel like the original tissue intended to be replaced. Moreover, no, or minimal, donor site morbidity would be involved. Last but not least, the procedure should be easy to perform. A special indication for soft tissue reconstruction is facial atrophy. Currently, in most of the cases, the only option for sufficient hemi facial reconstruction is free flap reconstruction.11e13 Theoretical alternatives are represented by the implantation of various materials such as gold, silver, wires, rubber, paraffin, silk, gutta-percha and ivory.14e16 In the middle of the last century, the use of free or solid silicon started a new era.17,18 In the last two decades, interest has focused more and more on liquids that can be easily injected.19 Historically, autologous fat transplantation was successfully performed at the end of the 19th century. The first scientific study on the survival of free fat transplantation was published in 1950 by Peer, and reported a survival rate of 50%.20 Vascularisation of fat droplets occurs after 4 days, before which the cells survive by diffusion. Several authors have published reports of permanent survival of fat grafts.21,22 Others have shown that traumatic handling of fat results in partial or complete resorption.23e25 Despite these reports, many plastic surgeons and other physicians still favour the transplantation of non-vital fat. In 1999, Coleman reported positive experiences with autologous facial fat transplantation in over 10 000 patients with follow ups of up to 8 years.26 Lipofilling has been presented primarily for the correction of cosmetic lesions or the reconstruction of minor soft tissue defects. Recently, however, the indication has been enhanced for larger soft tissue deficits or the
C. Hoehnke et al. improvement of facial proportions. Since aesthetic operations represent non-essential surgery it is very important to meet patients’ expectations in terms of the final result. For this reason, some surgeons take more than 200 photographs from various angles to get a feeling for the individual amount of fat required to augment local soft tissue deficits.26 Based on this problem we were looking for a method which would deliver reliable and reproducible data about the amount of soft tissue deficit, as well as provide a 3-D simulation of the optimal outcome. For many years, industrial settings have used a number of technologies for recording and visualising 3-D objects. Recently, these technical principles have been transferred into the medical field. The first applications have shown promising results.27 Three-dimensional visualisation of body parts enables a better understanding of the anatomy. In soft tissue defects, comparison studies with the normal counterpart or the ideal anatomy can deliver digital data on the defect or the tissue component to be reconstructed. With a background of successful experiences with 3-D planning in skull reconstruction and major facial defects27 we have taken advantage of the combination of both fat transplantation and 3-D assessment of the surface geometry by 3-D laser scanning in the reconstruction of this rare case of posttraumatic hemi facial atrophy. In order to obtain a model of the ideal right side of the face, 3-D scanning was used to mirror the non-injured left side resulting in a symmetrical face. The only problem lay in the nasal area, due to the deviation of the nose to the right. Based on the calculation of the local volume differences between the ‘ideal’ left side and the injured right side of the face, a topographic map of the soft tissue deficit was produced (Fig. 1). The total volume of the soft tissue deficit was primarily calculated to be 64 cc. Since the patient gained weight after the first treatment, the calculated soft tissue deficit before the second operation was higher than expected. The soft tissue deficit calculated for the third operation was about one-third of the injected fat volume in the second operation. This rate correlates with the rate of resorption. Consequently, in the last operation an over-correction of 21 cc was performed. The postoperative development of so-called ‘oil
Figure 3 Assuming that the mirrored left side of the face represents the ideal of the right side, (Fig. 3b) represents such a simulation in comparison to the preoperative situation in (Fig. 3a) and (Fig. 3c) where the final outcome is shown.
Minimal reconstruction of posttraumatic hemi facial atrophy cysts’ is a common problem in lipofilling. Since in our case this complication was seen particularly after locally injecting high volume fat deposits, we postulated that insufficient vascularisation and injection of traumatised or nonliving fat cells were the basic problems. Consequently, all injections were performed using a personal minimal traumatic technique involving three principles: (1) prior to all fat injections, local anaesthetics with adrenaline were administered, thus reducing local haematoma and postoperative pain which may result in high blood pressure and consecutive bleeding. (2) Fat was only deposited in micro channels created immediately beforehand, thus minimising trauma to the individual fat cells. (3) Three-dimensional fat deposition was achieved by a multi-layer ‘woven fat fabric’. Swelling returned to normal after about 8 weeks. Resorption was calculated at about 30%. Manual palpation of the reconstructed right side of the face in July 2005, 6 months after the last operation, showed only moderate differences compared to the left side of the face. Histological examinations of cheek biopsies taken in the third operation, 15 months after the first operation, confirmed the presence of vital fat tissue with only minimal fibrosis. With respect to facial symmetry, the latest photographs demonstrate significant improvement, especially in the mid face. Objective quality control was achieved by 3-D scanning technology. The final results of a precise mathematic comparison in surface geometry of facial soft tissue as well as volumetric differences support the subjective visual impression of facial symmetry improvement. Further correction could be achieved by an additional silicon implant in the chin, a septo-rhinoplasty and bilateral otoclisis. Since the patient was absolutely satisfied with the operative result, she declined any further intervention. The concept of using 3-D technology in facial reconstruction has multiple advantages. Primarily, the ideal final aesthetic outcome can be simulated by virtual reconstruction. Mathematic calculations deliver exact numbers in terms of volume deficits, enabling precise planning of soft tissue substitution especially in lipofilling, which ideally avoids any unnecessary corrections. Since autologous soft tissue reconstruction represents a dynamic process with periods of swelling as well as atrophy, quality control is required to achieve optimal results. Use of 3-D scanning has the advantage of reliable visualisation in soft tissue reconstruction without the limitations of harmful side effects. There are also new potential pathways in facial reconstruction. Although only one patient history is presented here as a case report, the results are absolutely convincing. The presented concept is even appropriate in HIV-related atrophy, rare syndromes and also in senile atrophy as well as wrinkles. In terms of treatment costs, this noninvasive concept of facial reconstruction has the significant advantage of reducing donor site morbidity and hospitalisation or absence from work. In conclusion, precise planning of minimally invasive hemi facial reconstruction and consecutive quality control can be achieved by using 3-D scanning technology in combination with complex lipofilling.
1143
References 1. Watzinger F, Wanschitz F, Wagner A, et al. Computer-aided navigation in secondary reconstruction of post-traumatic deformities of the zygoma. J Craniomaxillofac Surg 1997;25: 198e202. 2. Talmor M, Hoffmann LA, La Trenta GS. Facial atrophy in HIVrelated fat redistribution syndrome: anatomic evaluation and surgical reconstruction. Ann Plast Surg 2002;49:11e7. 3. Hodgkinson DJ. Facial atrophy in HIV-related fat redistribution: anatomic evaluation and surgical reconstruction. Ann Plast Surg 2003;50:328. 4. Gueganton C, Chavoin JP, Boutault F, et al. Treatment of facial lesions in Perry-Romberg and Barraquer-Simons syndromes: report of 12 cases. Ann Chir Plast Esthet 2000;45:436e51. 5. Kawano Y, Araki E, Arakawa K, et al. A case of progressive hemifacial atrophy with Pourfour de Petit syndrome which was successfully treated by stellate ganglion block. Rinsho Shinkeigaku 1999;39:731e4. 6. Mazzeo N, Fisher JG, Mayer MH, et al. Progressive hemifacial atrophy (Parry-Romberg-Syndrome). Case report. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1995;79:30e5. 7. Gambichler T, Kreuter A, Hoffmann K, et al. Bilateral linear scleroderma ‘‘en coup de sabre’’ associated with facial atrophy and neurological complications. BMC Dermatol 2001; 1:9. 8. Roddi R, Riggio E, Gilbert PM, et al. Progressive hemifacial atrophy in a patient with lupus erythematosus. Plast Reconstr Surg 1994;93:1067e72. 9. Kovacs L, Zimmermann A, Brockmann G. Three-dimensional recording of the human face with a 3D laser scanner. J Plast Reconstr Aesthet Surg 2006;59:1193e202. 10. Kovacs L, Zimmermann A, Wawrzyn H, et al. Computer aided surgical reconstruction after complex facial burn injuries opportunities and limitations. Burns 2005;31:85e91. 11. Siebert JW, Anson G, Longaker MT. Microsurgical correction of facial asymmetry in 60 consecutive cases. Plast Reconstr Surg 1996;97:354e63. 12. Longaker MT, Siebert JW. Microsurgical free-flap correction of severe hemifacial atrophy. Plast Reconstr Surg 1995;96: 800e9. 13. Lutz BS, Toussaint S, Wei FC. Bilateral lipoatrophy secondary to connective tissue panniculitis treated with two microsurgically transplanted latissimus dorsi muscles. Ann Plast Surg 1998;40: 302e7. 14. Beck J. Implantation method: Plastic surgery of the nose and ear. In: Loeb H, editor. Operative Surgery of the Nose, Throat, and Ear. St Louis: CV Mosby; 1917. p. 3. 15. Miller C. The limitations and use of paraffin in cosmetic surgery. Wis Med Recorder 1908;11:277. 16. Rogers B. A chronologic history of cosmetic surgery. Bull N Y Acad Med 1971;47:265. 17. Brown JB, Fryer MP, Randall P. Silicones in plastic surgery. Plast Reconstr Surg 1953;12:374e6. 18. Rees TD, Ashley FL. Treatment of facial atrophy with liquid silicone. Am J Surg 1966;111:531e5. 19. Hinderer UT, Escalona J. Dermal and subdermal tissue filling with fetal connective tissue and cartilage, collagen and silicone. Aesthet Plast Surg 1990;14:239e48. 20. Peer LA. Loss of weight and volume in human fat grafts. Plast Reconstr Surg 1950;5:217. 21. Moscona R, Ullmann Y, Har-Shai Y, et al. Free-fat injections for the correction of hemifacial atrophy. Plast Reconstr Surg 1989; 84:501e7. 22. Niechajev I, Sevcuk O. Long term results of fat transplantation: clinical and histological studies. Plast Reconstr Surg 1994;94: 496e506.
1144 23. Ellenbogen R. Invited comment on autologous fat injection. Ann Plast Surg 1990;24:297. 24. Ersek R. Transplantation of purified autologous fat: A 3-year followup is disappointing. Plast Reconstr Surg 1991;87:219e27. 25. Illouz YG. Fat injection: a four-year clinical trial. In: Hetter GP, editor. Lipoplasty: The Theory and Practice of Blunt Suction Lipectomy. Boston: Little Brown; 1990. p. 239e46.
C. Hoehnke et al. 26. Coleman SR. Fettunterspritzung (Lipostructure). In: LemperleA¨sthetische Chirurgie 1999;VI-2.1 [ecomed, 3. Erg Lfg]. 27. Kovacs L, Zimmermann A, Brockmann G, et al. Three-dimensional surface scanning of the facial region with the Minolta Vivid-910 3D Scanner: a survey of examining conditions, precision and accuracy using a dummy model. IEEE Trans Med Imaging 2006;25:742e54.