Treatment of enophthalmos using corrective osteotomy with concomitant cartilage-graft implantation

Journal of Plastic, Reconstructive & Aesthetic Surgery (2010) 63, 42e53

Treatment of enophthalmos using corrective osteotomy with concomitant cartilage-graft implantation Jing-Wei Lee* Division of Plastic Surgery, Department of Surgery, College of Medicine, National Cheng Kung University, and National Cheng Kung University Hospital, 138 Sheng-Li Road, Tainan 704, Taiwan Received 14 May 2008; accepted 21 August 2008

KEYWORDS Augmentation of soft-tissue volume; Concomitant cartilage grafting; Corrective osteotomy; Enophthalmos

Summary Post-traumatic enophthalmos is a relatively common problem following orbitozygomatic fractures. Bony-volume expansion and soft-tissue atrophication are considered the main aetiological causes of this condition. Although most surgeons are familiar with the treatment principles in this field, inadequate long-term results are frequently observed. The cardinal reason is due to overt volume deficits, owing to suboptimal reduction and the everexisting problem of soft-tissue atrophy. As such, it seemed logical that some treatment steps should be incorporated to increase the volume of orbital tissue. However, making fine adjustments to soft-tissue volume and orbital size during the same actual surgery is extremely difficult, if not impossible, which constitutes the biggest challenge in the treatment of enophthalmos. Based on the experiences from the management of seven patients with chronic enophthalmos (Group II), we could ascertain the average amount of the volume supplement required and were motivated to exploit a novel protocol of one-stage treatment for correction of disfiguring enophthalmos. In addition to the standard fracture-reduction methods, we use autologous, diced-cartilage graft to augment the orbital-tissue volume concomitantly for six consecutive patients (Group I) from 2004 to 2008. The actual quantities of inserted cartilage measured from 3.0 to 5.5 ml in total. An aesthetically and functionally satisfactory result is attained in every case thus treated, with only one patient exhibiting a minor degree of overcorrection (1 mm exorbitism). We thus advocate that this strategy is a viable option for preventing or rectifying late enophthalmos following severe orbitozygomatic fractures. ª 2008 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.

* Tel.: þ886 6 2353535x4275; fax: þ886 6 2766676. E-mail address: [email protected]

Post-traumatic enophthalmos is caused by either a relative deficiency of orbital tissue due to bony-volume expansion1 or an absolute reduction of tissue content resulting from fat loss or cicatricial contraction.2 Consequently, bonyalignment restoration and volume replenishment with

1748-6815/$ - see front matter ª 2008 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.bjps.2008.08.060

Concomitant grafting and osteotomy for enophthalmos grafting material are deemed the fundamental treatment modalities. Despite the seemingly straightforward therapeutic protocol, there still exist a multitude of articles elaborating upon the secondary correction of postoperative residual enophthalmos3e5 reflecting a significant incidence of treatment failure from primary surgeries. The reasons behind such a dismal prospect can be attributed to multiple causes. To begin with, there exist numerous technical pitfalls that may compromise the operative outcome. Improper reduction of the zygomatic complex,6 negligence of the floor and/or medial wall defect, deceptively agreeable contour of the lamina pyparacea in the initial stage, inadequate reshaping and positioning of the prosthetic metallic implant and insufficient release of the tethered soft tissue are among the common errors that may give rise to an unsatisfactory result. Although all these failings are well recognised and attentively avoided during surgeries, they just cannot be eradicated completely in clinical practice, and a degree of imperfect reduction always exists. Besides the extreme difficulties in restoring original bony anatomy, the decrease in soft-tissue volume poses another potential cause of enopththalmos. In the presence of the latter one, even an adequately reduced bony socket will not reproduce the pre-injury appearance. An additional manoeuvre to neutralise the discrepancy between orbital contents and bony volume is thus needed. Decreasing the size of the orbital cavity as suggested by Kawamoto7 or increasing the orbital-content volume with grafting material1,8 are the mainstay solutions. Yet, seldom were these two methods applied to the same patient at the same time. From the year 1992 to 2004, we had the experience of treating seven patients suffering from postoperative residual enophthalmos (Group II). In the face of dual problems involving both form and size, we elected to tackle each causal feature individually in a sequential manner, that is, to restore the skeletal alignment first and then to address the residual volume deficit by secondary grafting. In general, the amount of cartilage used for each case measured around 3.5e5.5 ml9 e with only one exception e a female patient who needed 7.0 ml of implant. The surgical result is quite encouraging, with an achievement of the left/right symmetry in five of them and only one case of minor exorbitism (overcorrection by 1 mm) and another one with residual enophthalmos (case no. 1, the earliest patient in Group II). However, such a two-staged surgical strategy implies added suffering and costs to the patients. In view of the high incidence of residual enophthalmos following primary surgery, we were motivated to exploit a more decisive and aggressive approach in the hope that we might be able to settle all the problems in a single session. A novel strategy combining prophylactic volume supplement with simultaneous corrective osteotomy procedure was thus formulated, and such a tactic has been applied in six consecutive cases of orbitozygomatic fractures with extensive inferioremedial wall destruction. From January 2001 to December 2007, there were 496 cases who suffered craniofacial fractures that were treated surgically in this hospital. Involvement of the oculofacial skeleton was recorded in 186 of them; a retrospective review of the chart and X-ray image was conducted in 153 cases, among which were eight patients who sustained

43 isolated medial orbital-wall disruption, 14 patients who had isolated orbital-floor fractures, 32 patients who had combined medial wall and orbital-floor fractures and 28 patients had varying degrees of residual enophthalmos after the primary surgery; 23 had their primary surgery done at this hospital, and five others were the cases referred from medical units elsewhere. A total of eight cases requested for surgical correction of disfigured enophthalmos in the above-mentioned period. The initial two cases (from 2001 to 2004) received the two-staged corrective operation, and then, after 2004, six ensuing patients underwent one-stage correction for both form and volume disorders, and they constitute the study group in this report (Group I). In addition to the ordinary reduction method, we use the autologous, diced, cartilage graft to augment the volume of the orbital content at the same time. The purpose of this study is to investigate the treatment outcome so as to elucidate the feasibility of such a new approach.

Material and methods A total of six patients enrolled in the main study group, which is designated as Group I. Four of the six patients are female, and all these four patients were referred to our clinic after failed primary surgeries. Two of them even had two surgical attempts prior to their visit. There are only two male patients in the group. One presented with a malpositioned zygoma with enophthalmos and came for a salvage surgery and the other patient was still in the acute trauma stage, and we deliberately carried out immediate grafting for him as a preventive measure to avoid late-onset enophthalmos. After the surgery, the patients were followed up periodically for the assessment of ophthalmic condition, including visual acuity, eyeglobe position, exophthalmometer record and extraocular muscle function. Image study using computerised tomography (CT) scan coupled with three-dimensional (3D) reformatting was routinely performed 3 months after the surgery. Another group of seven patients (Group II) who underwent the two-stage correction are also enlisted for comparison. The degree of enophthalmos, total amount of cartilage grafting and the treatment outcome are analysed to see if there are any differences among Group I and Group II.

Surgical techniques Group I: (single-stage operation) Under endotracheal general anaesthesia, the affected orbit is exposed through subciliary (skinemuscle flap) and upper blepharoplasty incisions. Marginal orbitotomy is used almost routinely in each case so that a non-obstructed view of the maxillary sinus can be obtained, which would expedite a safer and clearer dissection of the prolapsed orbital content that trespasses into the maxillary antrum. Corrective osteotomy is done at the fronto-zygomatic, zygomatico-maxillary, zygomatico-sphenoid and zygomatico-temporal junctions to free up the zygomatic complex.

44 Anatomical reduction of the bony framework and thorough release of the soft-tissue entrapment are then pursued. Metallic orbital-floor implants are used when the defect at the medial wall or orbital floor exceeds 1.5 cm in its largest dimension. The metallic plate should be fashioned conforming to the morphology of a normal orbit and then positioned precisely on the bony platform. Extensive subperiosteal dissection over all four walls is carried out to allow for unrestricted forward movement of the intra-orbital soft-tissue cone. Then we move on to harvest the rib cartilage from the right side of the chest to obviate the risk of heart injury. The cartilage is chopped into tiny pieces, and the volume is gauged using the Archimedes principle. A pre-estimated amount of cartilage is packed into the subperiosteal space. Prominent exophthalmos is invariably present at the end of the implantation, the eyelid skin is also stretched taut and a degree of lagophthalmos is witnessed as well.

Group II: (two-stage operation) The surgical skill required is almost the same as that in Group I, except that the treatment course is divided into two stages. Corrective osteotomy and repositioning of the mal-aligned skeleton is accomplished in the first-stage operation. Next, about 3e6 months later, a follow-up examination with Hertel’s exophthalmometer was undertaken. The degree of enophthalmos is used to estimate the total volume of diced-cartilage implantation.

Case report Case 1 (Group I) A 16-year-old girl was seen on consultation 3 months after suffering oculofacial trauma in a motorcycle accident. She was noted to have a comminuted right orbitozygomatic fracture and received surgical intervention on the fifth day of injury. Following the surgery, the patient had a sunkeneyeball deformity and binocular vertical diplopia, which worsened on upgaze. She was referred to the hospital after her functional and aesthetic problems failed to resolve. Evaluation revealed right-sided enophthalmos (5 mm lesser than the normal side) and hypophthalmos (Figure 1A and B). The orbital CT scan demonstrated a malpositioned zygomatic complex, with orbital-floor disruption and a medial-wall blow-out fracture as well (Figure 1C and D). At surgery, the zygoma was explored using subciliary, lateral eyebrow and superior buccal gingival incisions. Corrective osteotomy was performed to thoroughly mobilise the zygomatic complex, which was then repositioned and stabilised with miniplates. The infra-orbital nerve was dissected free from the entangling soft tissue to facilitate the latter to return to its original location. Then the orbital floor is restituted using a titanium orbital-floor implant after trimming and remoulding. Next we moved on to harvest the right-side costal cartilage via an inframammary incision. The cartilage is chopped into tiny pieces measuring 2e3 mm each. A total of 5 ml of diced cartilage were instilled into the orbital socket at all three quadrants surrounding the soft-tissue cone.

J.-W. Lee Seven months postoperatively, there was total resolution of the diplopia and the enophthalmos (Figure 2AeD). Case 2 (Group I) A 21-year-old lady was seen 4 months after having sustained an injury to the midface region in a car accident. Despite the acute-stage surgical attempt, the comminuted orbital floor and adjacent medial wall remained undercorrected. On examination, obvious hypoglobus and enophthalmos manifesting with deepened supratarsal sulcus were seen. The nose was flattened and skewed at the nasal-root region (Figure 3A). A three-dimensional CT imaging revealed a large orbitalfloor defect with concomitant involvement of the medialorbital wall (Figure 3B). The orbital contents and eyeglobe had herniated downwards into the maxillary antrum. At operation, the orbital floor was explored using a subciliary incision, and the medial orbital wall was exposed through the medial eyebrow incision. The orbital contents were returned back to its original space, and this dissection revealed a completely shattered orbital floor extending to the orbital apex region. We then applied a whole piece of costochondral graft to restitute the orbital floor. An additional 3.5 ml of cartilage chips was introduced into the retrobulbar region to push the eyeglobe anteriorwards. A segment of costochondral graft was used to augment the nose. Three months postoperatively, the patient had total resolution of her diplopia and complete correction of the enophthalmos (Figure 3C). Case 3 (Group I) A 24-year-old man suffered from a left comminuted orbitozygomatic fracture owing to a motor-vehicle accident. Open reduction with internal fxation using miniplates was performed 4 months earlier. The patient was referred to us with persistent facial deformity and enophthalmos. A CT scan confirmed the malpositioning of the orbito-zygomatic complex, which was displaced downwards and backwards, with resultant antimongoloid slant of the left eye (Figure 4A and B). There was a severe orbital-floor fracture with medial- and lateral-wall involvement. At operation, the orbito-zygomatic complex was exposed through upper and lower blepharoplasty incisions. Additional stab wounds at the sideburn area and the superior buccogingival incision were created to facilitate visualisation and osteotomy at the zygomatic arch and the maxillo-zygomatic junction. The zygomatic complex was refractured, repositioned and fixated with miniplates. The shattered medial wall and orbital floor was restituted with a titanium floor implant. Next, a costal cartilage graft was procured, diced and inserted into the retroequatorial orbital socket up to 4.5 ml in amount. The postopertiave course was uncomplicated. A satisfactory correction of the enophthalmos and malar asymmetry was confirmed, and the patient recovered completely from transient diplopia 5 weeks after the surgery (Figure 4C and D). Case 4 (Group I) A 33-year-old female sustained an orbitozygomatic fracture of the right side during a motorcycle accident 17 months prior to consultation.

Concomitant grafting and osteotomy for enophthalmos

45

Figure 1 Case no.1 of Group I: A 16-year-old woman with a chronic right cheekbone, orbital floor and medial-wall fracture. (A) Preoperative frontal view. (B) View from below: enophthalmos and depression in the zygomatic region clearly visible on the right side. (C) Preoperative CT (horizontal section), (D) Preoperative CT (coronal section).

She had been subjected to two surgical attempts without definite improvement. Physical examination revealed right-sided enophthalmos of 2 mm, hypoglobus and traumatic telecanthus (Figure 5A and B). A CT scan demonstrated medial-orbital wall and orbital-floor disruption and herniation of the intra-orbital content into the maxillary sinus (Figure 5C). At surgery, the prolapsed intra-orbital content was separated from the infra-orbital nerve and returned to the orbital cavity following marginal orbitotomy. Next, a piece of rib graft was used to replace the missing orbital floor. Later, we proceeded with medial canthopexy to relocate the laterally drifted medial eye angle to its proper position. Finally, we inserted 4.5 ml of cartilage graft into the retro-bulbar space to promote forwards shift of the eyeglobe position. At follow-up after 6 months, a CT scan imaging demonstrated fully replenished retro-ocular spaces with the cartilage graft (Figure 5D). A good position of the eyeglobe was maintained, and the telecanthus had also been corrected adequately (Figure 5E and F). Exophthalmometer measurement of the operated eye was recorded as 1 mm in excess of the fellow eye. The aesthetic appearance is perceived as quite acceptable to the patient and other viewers as well.

Case 6 (Group I) A 41-year-old male patient had a motorcycle accident, which resulted in a combined medial-wall and orbital-floor fracture on the left side. The bony buttress demarcating and supporting the medial and the inferior orbital walls had been crushed and displaced infero-medially (Figure 6A). He was operated upon 1 month after the event. In addition to the placement of a titanium orbital-floor implant, we electively introduced 3 ml of diced cartilage into the retroocular space to compensate for the potential volume discrepancy. A follow-up examination 4 months later revealed no disparity in the Hertel’s exophthalmometer readings between both eyes (Figure 6B and C). Case 7 in Group II A 28-year-old woman sustained a zygomatico-orbital fracture of the right side following a motorcycle accident. Open reduction and internal fixation were done 5 months prior to the visit. The patient came to us with the problem of dystopia and an overt enophthalmos deformity, evidenced by the deepened supratarsal sulcus and pseudoptosis (Figure 7A). We elected to proceed with a two-staged operation in this case. Corrective osteotomy with zygoma repositioning

46

J.-W. Lee

Figure 2 Case no.1 of Group I: Following corrective osteotomy and titanium orbital floor-implant insertion plus immediate volume augmentation using 5 ml ribecartilage grafting, the patient regained left/right symmetry at 7-month follow-up. (A) Postoperative frontal view (B) Postoperative worm eye’s view (C) Postoperative CT (horizontal section) (D) Postoperative CT (coronal section).

and orbital-floor repair was done as the first step of the procedure. Four months later, she underwent a secondstage surgery, in which 7 ml of diced cartilage was implanted retro-bulbarly to correct the residual enophthalmos problem (5 mm). There is no remnant discrepancy in the eyeglobe position at follow-up after 2 months (Figure 7B).

Result Five cases in Group I were referred from other institutes for secondary correction of mal-aligned skeletal and residual enophthalmos. The degree of preoperative enophthalmos ranged from 2 to 5 mm. The time lapse from the preceding surgery to consultation ranged from 3 to 17 months (mean 7.8 months). One case (Case 6) who suffered acute trauma was treated primarily with the same strategy. The patients were followed up for 2e35 months (mean: 15 months), and double vision was present in almost every patient shortly after the operation, which invariably resolved within 3 month following surgery. A total of

3.0e5.5 ml of cartilage were implanted into the retrobulbar space. There is no infection, visual impairment or long-standing diplopia in any of the cases. The Hertel’s exophthalmometer reading revealed no discernible discrepancy between the operated eye and the fellow normal eye in five of the six patients treated, including the one operated at acute trauma stage. There is only one patient (Case 4) having 1 mm of overcorrection (exorbitism), and yet the aesthetic result is perceived as quite pleasing by the patient herself and the treating physician as well. All patients expressed satisfactory response to the post-surgical aesthetics. No hyperphthalmos or hypoglobus were noticed. Telecanthus, epiphora or symptoms like sunken upper eyelid, widened supratarsal-fold width and pseudoptosis suggestive of enophthalmos were all absent in the follow-up period. The summaries of all cases are enlisted in Tables 1 and 2. The clinical data for the earlier group of seven patients (Group II) are shown as in Table 3. Those cases were subjected to the two-stage approach where volumisation was undertaken solitarily at a second session. It is observed that the orbital-volume change is closely correlated with the degree of enophthalmos. In contrast, for the Group I

Concomitant grafting and osteotomy for enophthalmos

47

Figure 3 Case no.2 of Group I: A 21-year-old woman with an old orbital-floor and medial-wall fracture on the right side, together with a depressed and deviated nose. (A) preoperative frontal view, evident hypophthalmos and enophthalmos with sunken upper eyelid deformity. (B) preoperative CT scan (coronal section), (C) postoperative frontal view. Fullness of the upper eyelid is restored following ribebone graft for the floor and 3.5 ml cartilage implantation.

Figure 4 Case no.3 of Group I: A 24-year-old man presented with an old displaced zygomatic fracture with extensive orbital-floor and medial-wall disruption. (A) Preoperative frontal view, (B) Preoperative basal view: downwards and backwards displacement of the zygomatic complex with antimongoloid slant. (C) Postoperative frontal view. (D) Postoperative basal view: result after corrective osteotomy, titanium orbital-floor implant and 4.5 ml ribecartilage graft, symmetric skeletal framework and eyeglobe position achieved.

Figure 5 Case no. 4 of Group I: A 33-year-old woman with right orbital-floor and medial-wall fractures. (A) Preoperative frontal view. Enophthalmos, hypophthalmos and telecanthus are evident on the right side. (B) Preoperative view from below. Retroposition of the eyeglobe is evident. (C) Preoperative CT scan. (D) Postoperative CT scan. (E) Postoperative frontal view. Following orbital-floor rib graft, 4.5 ml cartilage instillation and medial canthopexy, the patient regained aesthetically satisfactory appearance. (1 mm of overcorrection) (F) Postoperative basal view.

Concomitant grafting and osteotomy for enophthalmos

49

Figure 6 Case no. 6 of Group I: A 42-year-old man sustained left orbitozygomatic fracture with medial-wall and orbital-floor damage. He was submitted to open reduction, titanium orbital-floor-implant placement and immediate cartilage insertion with a volume of 3 ml. (A) Preoperative CT scan demonstrated medial-wall and -floor involvement. (B) Postoperative frontal view. Good symmetry of the eyeglobe position is obtained. (C) Postoperative CT scan. Note the distribution of the cartilage graft along the interior of the orbital socket.

subjects having had a one-stage restoration, such a correlation does not seem to hold; in other words, a graver enophthalmos may not consistently implicate a greater amount of volume requirement.

Discussion Disfiguring enophthalmos after orbitozygomatic trauma is a relatively common problem in clinical practice. For those suffering a low-velocity trauma resulting in a minor punchout fracture of the orbital floor, a simple obturator graft to seal off the defect would be sufficient. However, for those suffering a high-velocity impact resulting in bony-structural distortion, the reparative task becomes much more complicated. There are many adverse factors that may negate the surgical efforts and engender an imperfect treatment outcome.

One of the cardinal contributory factors is related to the intricate 3D configuration of the orbital socket. The crosssectional morphology starts as a round-shaped frame in the anterior portion (Figure 8A), then changes into a trapezoid form halfway inwards (Figure 8B) and is finally converted into a small triangular loop near the apical area (Figure 8C). As such, the medial flange of a metallic orbital floor implant should be bent in a steeply uprising slope to mimic the local morphological features of the natural orbit.10 Negligence to do this, such as by keeping the medial flange parallel to, rather than high above, the level of the inferior orbital rim, would definitely lead to an erroneous restorative result. Fractures straddling across the medial-orbital wall and the orbital floor are particularly troublesome. This is the site frequently violated in blow-out fractures,11 owing to its fragile architecture and its strategic location. However, the paper-thin bony shell may be displaced far medially and still look innocently intact, which in turn may trick the

50

J.-W. Lee

Figure 7 (A) Preoperative photograph of Case no. 7 in the earlier group (Group II) undergoing two-stage correction. (B) Postoperative photograph of the same patient. She had been treated with corrective osteotomy plus zygoma repositioning as the first step. Four months later, she underwent another surgery, in which 7 ml of diced cartilage was implanted retro-bulbarly to correct the residual enophthalmos problem (5 mm). There is no remnant discrepancy in eyeglobe position 2 months after the implantation procedure.

surgeons into accepting that aberrant situation as normality. The operators are working with an imaginary 3D image in their mind as the guide map; hence, a misjudgement is likely to occur. Theoretically, a prefabricated 3D model through computerised image processing and stereolithographic technology could be helpful in portraying the complex structural features,12 and a fit-in prosthesis thus generated might better facilitate the restorative work. However, the extra time and expense could be a costly trade-off, and, furthermore, such a prosthetic implant may not be easy to deploy properly. The task is made even more arduous due to the restricted access and limited exposure in the clinical setting. The fear of injuring the optic nerve and the eyeglobe poses an added obstacle to the exploratory act. In the face of all these unfavourable influential factors, it is not surprising that a deeply located medialwall fracture is oftentimes inappropriately treated. The failure in regaining original bony anatomy is a fact much more frequently encountered than generally thought or admitted.

Table 1

In addition to the practical difficulties in achieving exact anatomical bony reduction, there exists another confounding factor of actual soft-tissue decrement. Although bony-volume expansion has been widely accepted as a major aetiological cause for enophthalmos, it is by no means the only relevant parameter. A multitude of studies have illustrated that soft-tissue reduction does play a role.2,3 Post-traumatic fibrosis of the periocular adipose tissue is likely to occur, which will condense the fat component into a more compact and stiffer tissue block. Besides that, the intra-orbital content that herniates into the adjacent sinuses may undergo ischaemic necrosis due to strangulation of the nutrient vessel.13 In all these situations, a shrinkage in the soft-tissue bulk is apparently expectable, which in turn would provoke late enophthalmos. In light of this, volume manipulation should be considered an integral part of the treatment strategy. Since derangement of form and volume usually come together, a satisfactory result cannot be obtained with

Demographic data of Group-I patients.

Case no.

Sex

Age

Side

Previous surgery

Time from prevous operation

Presenting symptom

1 2 3 4 5 6

F F M F F M

16 21 24 33 35 42

R R L R L L

1 1 1 2 2 0a

3 months 4 months 4 months 1year and 5 months 11 months e

Enophthalmos, Enophthalmos, Enophthalmos, Enophthalmos, Enophthalmos, Enophthalmos,

a

Fresh case, without previous surgery.

hypoglobus, zygoma displacement hypoglobus, depressed and deviated nose zygoma displacement, antimongoloid slant hypoglobus, telecanthus hypoglobus, zygoma displacement hypoglobus, zygoma displacement

Concomitant grafting and osteotomy for enophthalmos

51

Table 2

Surgical data of Group-I patients.

Case no.

Pre op degree of enophthalmos

Pathology

Procedure

Cartilage graft Amount (ml)

Post-op degree of enophthalmos

1

5 mm

Medial and lateral wall and floor

5.0

0 mm

2

3 mm

Medial wall and floor

3.5

0 mm

3

3 mm

Medial and lateral wall and floor

4.5

0 mm

4

2 mm

Medial wall and floor

4.5

1 mma

5 6

3 mm 3 mm

Medial and lateral wall and floor Medial wall and floor

Corrective osteotomy þ Ti orbital floor implant Rib-bone graft for floor and nose augmentation Corrective osteotomy þ Ti orbital floor implant Rib-bone graft for floor and medial canthopexy Corrective osteotomy Corrective osteotomy þ Ti orbital floor implant

5.5 3.0

0 mm 0 mm

Ti - Titanium. a Exophthalmos of the treated eye.

a treatment method targeting at either item alone. For instance, if we rely on procedures that narrow down the bony socket and reduce the intra-orbital volume, we may end up with an asymmetrical facial architecture. On the other hand, if we merely fill up the intra-orbital space without fixing the bony framework, then the deformed skeleton certainly remains an unsightly stigma. Thus, neither of them employed alone would be capable of restoring the normal look. As such, a comprehensive treatment scheme addressing both form and volume should be pursued, which involves two conceptually diverse surgical measures. A critical issue that follows is related to the timing of the procedures. The surgeons are obliged to choose between the options of single- or two-staged surgery. A single-stage operation will save pain and time for the patients, and, yet, there is uncertainty regarding the exact volume adjustment that needs to be done. Therefore, we have to find a way to predict the approximate volumetric deficit right after bonystructure repair. Such a mission is rather difficult, since the volume will undergo a drastic change during skeletal repositioning, thus any clue based on the preoperative measurement or CT image analysis must be overturned and the calculated data invalidated. With such a degree of uncertainty, there is a substantial risk that we might misjudge the volume required and formulate an erroneous treatment plan. On the other hand, if we elect to adopt a staged approach by settling bony alignment first and then addressing the volumetric deficit with grafting later on, we may have better control over the final outcome, as evidenced in the Group II patients. During the second operation, the proper amount of grafting may be readily calculated from the disparity in exophthalmometer readings between both eyes.9 Such a two-stage approach would perhaps be advisable for those surgeons with lesser experience. Yet, the added suffering from a second operation is less welcomed by the patients and physicians as well. With increasing experiences in treating patients with established enophthalmos, we become aware of the average amount of graft required in such a clinical situation. The data stay mostly within a narrow range starting from 3.5

to 5.5 ml. Such information denotes the approximate amount of filler material we could safely insert, which serves as a valuable guideline in clinical practice. In a previous study done by Lee,9 the ratio between orbital-volume change versus that of eyeglobe-positional shift was estimated as being at about 1.37e1.5 ml of volume alteration for each 1 mm of eyeball movement. In a strict sense, such an equation is applicable only for cases in their chronic, stationary phases when the enophthalmos is well established. It is much less relevant for those in the acute trauma stage or for patients anticipating major skeletal framework reshuffling. Nevertheless, such information is still of practical value in the surgical scheme, giving us a basic idea regarding the gross effect following implant insertion. For instance, when 5 mL of graft is inserted, there would be a corresponding forward movement of the eyeglobe of about 3.5 mm in extent. In those having a predestinated enophthalmos with a scale of 2e 4 mm, such an advancement would lead to a final result ranging from 1.5 mm of exorbitism to 0.5 mm of residual enophthalmos. Such a terrain of outcome is deemed aesthetically acceptable. Apparently, the safety range for a desirable reconstruction is forgivingly wide, and a reasonably decent result can be readily attainable.

Table 3

Data for Group-II patients.

Case Sex Age Side Pre op Graft Post-op degree no. degree of (ml) of enophthalmos enophthalmos (mm) (mm) 1 2 3 4 5 6 7

M M M F M M F a

28 54 32 28 40 24 28

R R R R L L R

4 3 3.5 3 2 3 5

Exophthalmos of the treated eye.

3 5.5 5 4.5 3 4.5 7.0

2 1a 0 0 0 0 0

52

J.-W. Lee

Figure 8 A cross-sectional view of the orbital socket at different levels from the front to the back. (A) At the orbital outlet, the skeletal framework assumes a round-shaped configuration. (B) Proceeding halfway inwards, the bony cage transforms into a trapezoid shape. (C) While approaching the apical region, the orbital socket turned into a triangular loop.

Another source of reference comes from the experiences of decompression surgery for Graves’ exophthalmopathy.14 Stabile15 has measured the increase in orbital volume obtained by orbital-wall removal in dried skulls. The volume changes following unroofing of the paranasal sinuses are recorded as follows: 6 ml for the ethmoid sinus, 6e7 ml for the medial portion of the maxillary antrum, 1e 2 ml for the lateral portion of the maxillary antrum and 2 ml for the lateral orbital wall. Such a measurement denotes the maximal volume change that may occur when the specific component is involved in a blow-out fracture. This data can also be used as a reference to extrapolate the approximate amount of volume replenishment needed. Specifying the required amount of cartilage in a given patient is essentially an empirical task. The range and severity of damage are taken as the base of estimation. With the bony involvement of two compartments in modest degree (medial-wall displacement lesser than 3 mm), we elect to insert 3.0e4.0 ml, with damage involving two

compartments in a severe degree (medial-wall displacement more than 3 mm) or all three compartments, we apply 4.5e5.5 ml of cartilage implant. A total of 3.0e5.5 ml of cartilage has thus been applied in our series. The tension of the eyelid and the protruding status of the eyeglobe are monitored and serve as a reference indicators to proceed with or rein back the implantation manoeuvre. One potential criticism of our technique is that we should seek for anatomical reduction rather than resorting to volume augmentation in the first place. However, it becomes increasingly clear that actual loss of soft-tissue content really exists in such group of cases. In light of this, an overcorrection instead of anatomical restoration would be more appropriate. One possible measure is to mould the metallic implant in an exaggerated fashion to decrease the effective bony volume and drive the soft-tissue contents forwards. However, there are several inherent drawbacks with such a manoeuvre. First, the rigid sharp edge of the metallic implant, if remoulded excessively inwards, is

Concomitant grafting and osteotomy for enophthalmos prone to impinge upon the delicate structures in the apical area, causing neurovascular damage or curbing the extraocular musculature. Second, the concern of the above hazard would prohibit the surgeon from advancing the implant too deeply, thus leaving a leakage port in the innermost area, allowing soft-tissue content to slip out over the edge of the implant. Third, if the bending is done too aggressively, there is the threat of displacing the eyeglobe eccentrically and causing diplopia.10 In other words, attempts to adjust the intra-orbital volume through metallic-implant manipulation is plagued with troubles on account of its poor manoeuvrability. In contrast, with the highly malleable character of the diced cartilages, we could readily insert the filler into the retro-bulbar region with little apprehension. The filler tends to slip into the suitable space automatically in relatively even distribution, almost like the instillation of a semiliquid material into a glass, thereby keeping the tissue in the orbit well balanced as it is moved in an anterior direction. It appears quite puzzling why we do not encounter many incidences of overcorrection or proptosis. A probable explanation is that the soft-tissue contents are somewhat compressible, thus when the indwelling cartilage raises up the total volume of orbital contents, the intraconal pressure and the soft-tissue density increases correspondingly (an unpublished work from our study), which may partly counteract the forward-thrusting effect of the implantation and reduce the chance and degree of offending proptosis. From a practical point of view, a slight exorbitism would be more youthful looking and aesthetically agreeable than the enophthalmic deformity. In other words, we would rather err on the excessive side than on the deficient side. Therefore, even if modest overcorrection does occur, it would not be much of a problem, and as a matter of fact, it has not been encountered in our group of cases. Another debatable issue could be related to the selection of surgical incisions. A coronal incision and lower eyelid swing have the merits of avoiding visible scars in the facial region, hence providing a less-disturbed milieu for intra-operative monitoring of the oculo-orbital balance. Yet the attached price from coronal incision is the extensiveness of dissection with increased blood loss, longer operation time, scalp sensory deficit and alopecia. Lower eyelid swing may suffer from the potential sequelae of granulation-tissue overgrowth, posterior lamellar contracture and resultant entropion or sclera show on the lower pole. Therefore, the upper blepharoplasty incision and subciliary access are employed in most of our cases, with the added alternative of eyebrow incision on three occasions. We observed that such an approach is less strenuous, providing adequate exposure and enjoying an acceptable scar quality. An argument may exist as to the choice of filler material we should use. Autologous fat, fascia, bone graft or alloplastic materials are enlisted among the potential alternatives. Despite the advantage of absolutely no donor sacrifice, the use of alloplastic material can be plagued by problems such as infection, chronic inflammation, capsule contracture or migration and is considered an unreliable option. The fat and fascia will have variable rates of survival, and iliac-bone graft also suffers a high incidence

53 of absorption; thus relapse is frequently a problem. Calvarial bone graft, although allegedly having a low resorption rate, is too rigid to reshape and process, and the risk of soft-tissue impingement or incarceration is heightened. In conclusion, the diced cartilage is superior in terms of durability, manoeuvrability and safety and works well in such a clinical situation. With the judicious use of a combined approach, including prophylactic cartilage implantation and corrective osteotomy, we might be able to solve both shape and size problems of complicated oculofacial fractures in a single session, with effective rectification and prevention of late enophthalmos.

Conflict of interest The author affirms that there is no conflict of interest involved in the present work, and there is no funding from any party or organisation contributory to this study.

References 1. Manson PN, Grivas A, Rosenbaum A, et al. Studies on enophthalmos: II. The measurement of orbital injuries and their treatment by quantitative Computed Tomography. Plast Reconstr Surg 1986;77:203e14. 2. Bite U, Jackson IT, Forbes GS, et al. Orbital volume measurements in enophthalmos using three-dimensional CT imaging. Plast Reconstr Surg 1985;75:502e7. 3. Nishi Y, Kiyokawa K, Watanabe K, et al. A surgical treatment of severe late posttraumatic enophthalmos using sliced costal cartilage chip grafts. J Craniofac Surg 2006;17:673e9. 4. Chen CT, Huang F, Chen YR. Chang Gung Med J 2006;29: 251e61. 5. Longaker MT, Kawamoto HK. Evolving thoughts on correcting posttraumatic enophthalmos. Plast Reconstr Surg 1998;101: 899e906. 6. Pearl RM. Prevention of enophthalmos. A hypothesis. Ann Plast Surg 1990;25:132e3. 7. Kawamoto HK. Late posttraumatic enophthalmos: a correctable deformity? Plast Reconstr Surg 1982;69:423e30. 8. Matsuo K, Hirose T, Furuta S, et al. Semiquantitative correction of posttraumatic enophthalmos with sliced cartilage grafts. Plast Reconstr Surg 1989;83:429e37. 9. Lee JW. Preplanned correction of enophthalmos using diced cartilage grafts. Br J Plast Surg 2000;53:17e23. 10. Lo AK, Jackson IT, Ross JH, et al. Severe orbital floor fractures: repair with a titanium implant. Eur J Plast Surg 1992;15: 35e40. 11. Burm JS, Chung CH, Oh SJ. Pure orbital blowout fracture: new concepts and importance of medial orbital blowout fracture. Plast Reconstr Surg 1999;103:1839e49. 12. Holck DEE, Boyd EM, Ng J, et al. Benefits of stereolithography in orbital reconstruction. Ophthalmology 1999; 106:1214e8. 13. Chien HF, Wu CH, Wen CY, et al. Cadaveric study of blood supply to the lower intraorbital fat: etiologic relevance to the complication of anaerobic cellulitis in orbital floor fracture. J Formos Med Assoc 2001;100:192e7. 14. Stranc M, West M. A four-wall orbital decompression for dysthyroid orbitopathy. J Neurosurg 1988;68:671e7. 15. Stabile JR, Trokel SM. Increase in orbital volume obtained by decompression in dried skulls. Am J Ophthalmol 1983;95: 327e31.