ORIGINAL STUDY
Pediatric Orbital Blow-Out Fractures: Surgical Outcomes Jamie Y. Ng, MRCS Ed,* Sundar Gangadhara, FRCS,* Wong Wanling, PhD,Þ and Shantha Amrith, FRCOphth*
Purpose: The objective of this study was to evaluate the features, timing of intervention, complications, and outcomes of patients who underwent surgery for pediatric orbital blow-out fractures. Design: This was a retrospective case review studying all cases of pediatric orbital blow-out fractures that underwent surgical intervention from 2000 to 2009 in a tertiary ophthalmic center in Singapore. Methods: Case notes review of all cases of pediatric orbital blow-out fractures repaired surgically was carried out for demographics, time to intervention, and outcome. How the various factors may affect the outcome was statistically analyzed. Results: A total of 23 patients were reviewed. There were 21 patients (91.3%) with floor fractures, 1 (4.3%) with medial wall fracture, and 1 (4.3%) with combined medial wall and floor fracture. Of the 21 cases of floor fracture, 20 cases had radiological images available for review, 17 (85%) were linear, and 3 (15%) were open. Patients with open fractures were older with a mean age of 14.0 years as compared with 12.6 years in those with linear fractures. The mean duration from injury to intervention was 13.0 days. Preoperatively, 14 (60.9%) had diplopia, but only 4 cases (17.4%) had diplopia in extreme gaze at the last follow-up after surgery (18 months). Outcome was not affected by age, duration between injury and intervention, type of fracture, and nature of implants used. Conclusions: Our study demonstrated that pediatric orbital blow-out fractures had good outcomes. It might also suggest the possibility of not having to manage all linear fractures as a surgical emergency. Key Words: blow-out fracture, white eye, pediatric (Asia-Pac J Ophthalmol 2012;1: 265Y269)
T
he management of orbital blow-out fractures has always been a topic of controversy, and pediatric fractures are certainly not as well studied as adult fractures. The greater elasticity in pediatric bones as compared with adults results in greenstick deformity that manifests as trapdoor fractures.1Y7 Soft tissues such as extraocular muscle, intermuscular septa, and fat may become entrapped, resulting in significant restriction in eye movement. This might be without suggestive periocular signs of trauma and hence known as ‘‘white-eyed blow-out fracture’’ (Fig. 1), which is often missed.8 The purpose of our study was to evaluate the clinical, radiological features, and timing of intervention and correlate them with the surgical outcomes of pediatric orbital blow-out fracture repair.
that underwent surgical intervention by 2 senior oculoplastic surgeons from January 2000 to December 2009 was conducted. The age of 16 years was chosen as the cutoff age for inclusion into the study based on our institutional guidelines. There were a total of 23 patients. All patients underwent computed tomography (CT) scans. Axial, coronal, and in some cases sagittal reconstructions in both bone and soft tissue windows were read both by a radiologist and as well as the oculoplastic surgeon. Clinical evaluation included a detailed history and complete ophthalmologic examination including mechanism of injury, symptoms of nausea and vomiting, associated ocular injuries, diplopia in primary as well as different positions of gaze, extraocular eye movements, enophthalmos, and infraorbital paresthesia. Duration from injury to ophthalmic referral and duration from injury to intervention were also noted. Surgical factors such as the approach to the orbit, type of implant (according to patient’s and parent’s choice), and complications were examined. The outcome of surgery was measured by noting the improvement in diplopia and extraocular eye movements, resultant enophthalmos, and any residual complications. All patients received prophylactic systemic antibiotics (amoxicillin and clavulanate) postoperatively. Linear fractures (trapdoor), as the name suggests, are defined as orbital floor fractures that extend anteroposteriorly, allowing the herniation of orbital contents and entrapping them. Open fractures are usually hinged medially and have a gaping defect that allows the orbital contents to herniate through. Minimal diplopia was defined as diplopia present only in extreme gaze, moderate as diplopia present from 15 to 25 degrees from central fixation, and severe when diplopia was present at within 15 degrees from central fixation. The extraocular eye movements were likewise graded.
RESULTS All surgeries were performed by either of 2 senior oculoplastic surgeons in our center, and all surgeries were performed through a transconjunctival route, most of them not requiring cantholysis. The indications for surgery included clinical or radiological evidence of entrapment of tissue, large fracture greater than 50% of orbital floor, diplopia (any moderate or severe diplopia and/ or restriction of eye movements in Q1 positions of gaze), and persistent and significant enophthalmos greater than 2 mm.
MATERIALS AND METHODS With the approval of the institutional review board, a retrospective case review of all pediatric orbital blow-out fractures From the *Department of Ophthalmology, National University Health System, Singapore; †Singapore Eye Research Institute, Singapore. Received for publication April 16, 2012; accepted May 15, 2012. The authors have no conflicts of interest to declare. Reprints: Jamie Y. Ng, MRCS Ed, Department of Ophthalmology, National University Health System, 1E, Kent Ridge Rd, Tower Block, Level 7, Singapore 119228. E-mail: [email protected]. Copyright * 2012 by Asia Pacific Academy of Ophthalmology ISSN: 2162-0989 DOI: 10.1097/APO.0b013e31825f 8976
Asia-Pacific Journal of Ophthalmology
&
FIGURE 1. A child with a right white-eyed blow-out fracture of the orbit with severe limitation in up-gaze.
Volume 1, Number 5, September/October 2012
www.apjo.org
Copyright © 2012 by Asia Pacific Academy of Ophthalmology. Unauthorized reproduction of this article is prohibited.
265
Asia-Pacific Journal of Ophthalmology
Ng et al
Patient demographics are shown in Table 1. The mean age was 12.9 (SD, 2.6) years (range, 5Y16 years). Sixteen had linear fractures, and 4 had open fractures. The group with open fractures was slightly older than the group with linear fractures (mean, 14.0 vs 12.6 years). This was not statistically significant. The patients were predominantly male (n = 21; 91.3%). Of the 23 children operated on, the majority were Chinese (n = 14; 60.9%) followed by 7 Malays (30.4%) and 2 Indians (8.7%). There was thus a slight overrepresentation of Malays compared with the racial profile of Singapore, reported to consist of 74.2% Chinese, 13.4% Malays, 9.2% Indians, and 3.2% others. The mean duration of postoperative follow-up was 9.6 (SD, 8.0) months (range, 1Y42 months; median, 8.0 months); 3 cases defaulted follow-up after 1 month. Vasovagal symptoms were
TABLE 1. Demographics (N = 23) Demographics Age, mean (SD), y Linear factures Open fractures Sex, n (%) Male Female Race Chinese Malay Indian Follow-up, mean (SD), mo Vasovagal symptoms, n (%) Yes No Time from injury to ophthalmic referral, n (%) Same day 1 d postinjury 2 d postinjury 3 d postinjury 4 d postinjury 5 d postinjury Associated injuries, n (%) Yes No Laterality, n (%) Right Left CT findings,* n (%) Floor fracture Medial wall Combined Type of floor fracture* Linear Open Nature of implants (n = 21), n (%) Medpor Bioresorbable Supramid
Range 12.9 (2.6) 12.6 (2.9) 14.0 (1.1) 21 (91.3) 2 (8.7) 14 (60.9) 7 (30.4) 2 (8.7) 9.6 (8.0) 5 (22.7) 17 (77.3)
12 (52.1) 5 (21.7) 3 (13.0) 1 (4.4) 1 (4.4) 1 (4.4) 15 (65.2) 8 (34.8) 11 (47.8) 12 (52.2)
www.apjo.org
1.0Y42.0
Volume 1, Number 5, September/October 2012
TABLE 2. Mechanism of Injury (N = 23) n (%) Assault Sports Domestic accidents Pedestrians in road traffic accident
8 (34.8) 8 (34.8) 4 (17.4) 3 (13.0)
reported in 5 patients (22.7%). Majority (73.8%) was referred for ophthalmic evaluation within a day after injury, and the remaining within 5 days from injury. Hence, the mean time from injury to an ophthalmic referral was 1.0 day. At presentation, 15 patients (65.2%) had associated injuries. The most commonly associated injury was periorbital hematoma (n = 10; 41.7%); others included 3 upper eyelid lacerations (12.5%), 2 commotio retinae (8.3), and 1 case of microscopic hyphema (4.2%). Table 1 also describes the mode of injury. Assaults and sports injuries were equally common followed by domestic accidents and road traffic accidents. Sports activities included soccer (n = 4), baseball (n = 2), and softball (n = 2). Table 2 summarizes the fracture characteristics as seen in the CT, the time of intervention, and the outcome. The CT scans were reviewed by radiologists and ophthalmologists. For 1 case of floor fracture, the scans were done elsewhere and not available for review. Of the 20 cases that were isolated floor fractures, 17 (85.0%) were linear (Fig. 2), and 3 (15.0%) were open. Three (13.6%) of 22 scans were reported by radiologists as without any orbital fractures, but because of a strong clinical suspicion and subtle changes in CT scan suggestive of a fracture, surgery was performed, and the presence of fracture was confirmed in all 3. An example is shown in Figure 3. A detailed breakdown of duration from injury to surgery is given in Table 3. The median duration from injury to time of intervention was 13.0 (SD, 17.0) days (range, 0Y74 days; median, 10.0 days). Ten cases had surgery within 1 week of injury, and 6 others had surgery between 1 and 2 weeks after injury. The indication for surgery in the case that was operated 74 days after injury was residual enophthalmos. Mean time from injury to surgical intervention for linear fractures was 13.7 days and for open fractures was 14.8 days. Patients were evaluated for preoperative and postoperative infraorbital paresthesia, enophthalmos, and diplopia. Four patients (17.4%) had infraorbital anesthesia preoperatively, and it persisted in 2 cases beyond 1 month after surgery. Table 4 shows the comparison between preoperative and postoperative diplopia. Only 14 patients (60.9%) had preoperative diplopia, among whom 3 were present in primary position. None of
20 (90.9) 1 (4.5) 1 (4.5) 17 (85.0) 3 (20.0) 15 (71.4) 4 (19.0) 2 (9.5)
*Imaging films were not available for 1 subject.
266
5.0Y16.0 5.0Y16.0 12.0Y15.0
&
FIGURE 2. Computed tomographic imaging showing a linear right orbital floor fracture in the coronal cut with entrapment of orbital contents. * 2012 Asia-Pacific Journal of Ophthalmology
Copyright © 2012 by Asia Pacific Academy of Ophthalmology. Unauthorized reproduction of this article is prohibited.
Asia-Pacific Journal of Ophthalmology
&
Volume 1, Number 5, September/October 2012
FIGURE 3. Computed tomographic imaging showing a left linear orbital floor fracture that was initially missed by the radiologist.
the patients had poor vision limiting the assessment of diplopia. Overall, 4 patients (17.4%) had residual mild diplopia after surgery at the last follow-up (18 months), and in all cases diplopia was far
Pediatric Orbital Blow-Out Fractures
less symptomatic than at presentation. None had significant residual diplopia that affected their daily activities. Overall, the diplopia in primary position and up-gaze showed significant improvement (P = 0.006 and 0.014, respectively). Preoperative exophthalmometric readings were not included in the analysis as many patients had proptosis due to the orbital hematoma and edema, and a considerable number underwent surgery before it settled. Two patients (8.7%) had mild residual enophthalmos (3 mm) postoperatively, but both were asymptomatic (Table 5). Residual diplopia in any gaze position in relation to time from injury to intervention was analyzed. There were no significant differences in this final outcome in the early (G48 hours) and late (92 weeks) intervention groups (P = 0.876). Patients who presented with diplopia regardless of severity and positions of gaze were operated on at a mean duration of 7.2 (SD, 8.2) days (range, 0Y26 days), as compared with children without diplopia at presentation [mean, 22.1 (SD, 23.0) days; range, same day to 74 days]. Patients with any
TABLE 3. Fracture Characteristics, Time of Intervention, and Outcome Case Number 1* 2 3 4
Type of Fracture
Image (CT)
TrapdoorVjunction of medial wall and floor Trapdoor
A, C, S A, C, S
Trapdoor Open fracture of floor and medial wall Trapdoor Trapdoor
5 6 7
TrapdoorVjunction of medial and floor Open Trapdoor Trapdoor
8 9* 10 11 12 13 14
A, C, S A, C, S A, C, S A, C A, C A, C, S A, C A, C
15 16 17
Medial wall A, C, S Open A, C Trapdoor A, C TrapdoorVjunction of medial A, C, S wall and floor Scans not available for review Trapdoor A, C, S Trapdoor A, C
18*
Trapdoor
A, C
19 20
Trapdoor Trapdoor
A, C, S A, C, S
21 22 23
Open Trapdoor TrapdoorVjunction of medial wall and floor
A, C A, C A, C
Injury to Surgery Time
Outcome
Entrapment of fat
0
Good
Entrapment of fat and muscle sheath Entrapment of fat IR muscle inside fracture
0
Diplopia in extreme gaze
0 0
Diplopia in extreme gaze Good
Entrapment of IR muscle Entrapment of fat and muscle sheath Entrapment of IR muscle
1 1
Good Good
1
Diplopia in extreme gaze
IR muscle inside the fracture Entrapment of muscle sheath Entrapment of fat and muscle sheath MR inside the fracture IR muscle inside fracture Entrapment of fat Entrapment of fibrous connective tissue
1 4 5
Good Good Good
9 10 11 12
Good Mild enophthalmos Good Good
13 13 15
Good Mild enophthalmos Diplopia in extreme gaze
16
Good
20 26
Good Good
28 40 74
Good Good Good
Image Findings
Entrapment of fat Entrapment of fat and muscle sheath Entrapment of fat and muscle sheath Entrapment of IR muscle Entrapment of fat and muscle sheath IR muscle inside the fracture Entrapment of fat Entrapment of fibrous connective tissue
*Missed by radiologist. A indicates axial; C, coronal; S, sagittal; IR, inferior rectus; MR, medial rectus.
* 2012 Asia-Pacific Journal of Ophthalmology
www.apjo.org
Copyright © 2012 by Asia Pacific Academy of Ophthalmology. Unauthorized reproduction of this article is prohibited.
267
Asia-Pacific Journal of Ophthalmology
Ng et al
&
Volume 1, Number 5, September/October 2012
TABLE 4. Duration Parameters Range Duration from injury to surgery, no. cases (%) G48 h Q48 h to G1 wk Q1 wk to G2 wk Q2 wk Mean time from injury to surgical repair, mean (SD), d Overall Presented with initial diplopia that resolved With residual postoperative diplopia Linear fractures Open fractures
FIGURE 4. Preoperative photograph showing right eye limitation in up-gaze in a child with right linear floor fracture.
8 (34.8) 2 (8.7) 6 (26.1) 7 (30.4)
13.0 (17.0) 7.2 (8.2)
0Y74
4.0 (7.3) 13.7 (19.5) 14.8 (11.8)
residual diplopia after surgery were intervened at an average of 4.0 days (same day to 15 days) as compared with 7.2 days for those who presented with diplopia initially that resolved after surgery. There was no statistical difference in the outcome of these 2 groups of patients (P = 0.375). Postoperative diplopia was also not found to be significantly influenced by
TABLE 5. Comparison of Preoperative and Postoperative Diplopia and Enophthalmos Preoperative, Postoperative, n (%) n (%) Diplopia Yes No Primary position Yes No Up-gaze None Mild Moderate Severe Down-gaze None Mild Moderate Severe Horizontal None Mild Moderate Severe Enophthalmos at last visit after surgery† Yes No
DISCUSSION P* 0.006
14 (60.9) 9 (39.1)
4 (17.4) 19 (82.6)
3 (13.0) 20 (87.0)
0 (0.0) 23 (100.0)
11 (47.8) 6 (26.1) 2 (8.7) 4 (17.4)
20 (87.0) 3 (13.0) 0 (0.0) 0 (0.0)
20 (87.0) 2 (8.7) 0 (0.0) 1 (4.3)
23 (100.0) 0 (0.0) 0 (0.0) 0 (0.0)
20 (87.0) 1 (4.3) 1 (4.3) 1 (4.3)
22 (95.7) 1 (4.3) 0 (0.0) 0 (0.0)
0.006
0.014
0.233
0739
V
2 (8.7%) 21 (91.3%)
V
*P value from Fisher exact test. †Preoperative measurements not included as some patients had proptosis due to hematoma and edema.
268
www.apjo.org
age (P = 0.596), fracture type (P = 0.861), or implants used (P = 0.999). Of the 23 cases, 22 had implants, but the type of implant was unknown in 1 case. One patient did not require an implant, as there was no significant residual defect after the inferior rectus entrapment was released. Porous polypropylene (MedporBarrier Sheet; Porex Surgical Inc, College Park, Ga) (15; 71.4%) was the most commonly used implant, followed by bioresorbable PolyMax (Synthes, Oberdorf, Switzerland) (4; 19.0%). Supramid (Ethicon, Somerville, NJ) was used in 2 patients (9.5%) who underwent surgery before 2002. These findings are indicated in Table 1. We had only 1 case with early postoperative retrobulbar hemorrhage presenting with optic neuropathy hours after surgery in a patient with large medial wall fracture, which was repaired with Medpor implant. This was managed conservatively with intermittent cold compression and intravenous methylprednisolone, and the patient recovered well without residual visual deficit. Other complications from the surgery such as infection, vision loss, or implant-related complications were not seen in any case. Hence, the overall rate of complication was 4.3%.
Current recommendations for surgical intervention in orbital blow-out fractures include muscle and soft tissue entrapment resulting in significant diplopia, enophthalmos of more than 2 mm, and orbital defects involving greater than 50% of either the medial wall or the floor. The urgency of management of adult and pediatric cases has always differed. Surgery in the pediatric group has been conducted more urgently in view of the greater potential for entrapment resulting in muscle ischemia and fibrosis.1,2,4,9 It has also been demonstrated that even with early intervention, limitation in eye movements may be present in the early postoperative period because of ischemia, edema, and hemorrhage, but they tend to recover completely in weeks to months.10 Most studies in the literature on pediatric blow-out fractures are limited by the small numbers similar to our study, making statistical analysis impossible. The ischemia and necrosis of the trapped extraocular muscle (inferior rectus in most cases) has not been clearly demonstrated. Various studies suggest early intervention in pediatric population, but the conclusions derived about timing of surgical intervention versus results are anecdotal. A magic number for trauma to surgical intervention time may not exist in pediatric blow-out fractures because the urgency of intervention may be dependent on individual case’s clinical and CT scan findings. Our study suggests that there are no significant differences in the final outcome in the early (G48 hours) and late (92 weeks) intervention groups. This may indicate that the diplopia outcome may be dependent on the severity of the initial injury and fracture. Although a 1-week interval between injury and intervention is acceptable in most cases, one should always take the age of the patient, clinical signs, and the nature of entrapment into account while managing pediatric patients with blow-out fractures. It appears that our physicians in the frontline are able to recognize the clinical features of pediatric orbital fractures and * 2012 Asia-Pacific Journal of Ophthalmology
Copyright © 2012 by Asia Pacific Academy of Ophthalmology. Unauthorized reproduction of this article is prohibited.
Asia-Pacific Journal of Ophthalmology
&
Volume 1, Number 5, September/October 2012
FIGURE 5. Early postoperative photograph demonstrating full range of movement in up-gaze in the same child.
are able to act quickly as evidenced by referrals within a day to the oculoplastic team in more than three fourths (73.8%) of the patients. It will be useful to do a prospective data collection about the flow of such patients within the hospital to determine where the delay is occurring, so that the hospital processes can be modified to shorten the duration between trauma and surgery. Although the follow-up period in 3 of our cases was only 1 month, all of them had no diplopia or enophthalmos at their last visit, and therefore, our statistical analysis was not adversely affected by the short follow-up. Computed tomographic scans have been widely accepted and useful in the diagnosis of orbital fractures. Orbital fractures have been reportedly missed radiologically on CT scan,1,7 and this was evident in our study as well. White-eyed blow-out fractures have subtle signs both physically and radiologically. Hence, all scans should be carefully evaluated and correlated to clinical findings before ruling out the diagnosis of a fracture. Children with trapdoor fractures tend to have nausea and vomiting.1,4,11 It has been related to the oculocardiac reflex, and such symptoms should alert all ophthalmologists, especially in young children who may not be able to express symptoms of diplopia. Our study reports a rate of vasovagal symptoms of 22.7%, which is significantly lower than 75% as reported by Lane et al.11 Our study, being retrospective, may not have captured such systemic symptoms because of improper documentation in the case records. Although no implant-related complications were seen in this series, bioresorbable implants are probably better in pediatric orbital blow-out fractures. It is ideal for linear fractures where the defect tends to be small and a long-term structural support is not required. The parents also like the idea of their children not having to live with permanent implants in their bodies. However, in large fractures, it is still not clear if the bioresorbable implants afford the kind of structural support necessary after complete resorption; therefore, a more permanent implant such as Medpor or titanium may be necessary to support tissues. We acknowledge that our study results are limited by the small sample size. The other limitation is that we chose to study diplopia as the outcome measure rather than restriction of eye movement. Although the decision to intervene surgically was based on motility restriction combined with forced duction test and CT scan findings, not all children complained of diplopia. It was difficult to quantify the motility restriction because of noncooperation from children partly because of the pain associated with the eye movements.
* 2012 Asia-Pacific Journal of Ophthalmology
Pediatric Orbital Blow-Out Fractures
Our study confirms that pediatric blow-out fractures have good outcome without significant complications after surgical intervention and suggests that not all cases need to be intervened as an emergent condition. Figures 4 and 5 demonstrate an example of good postoperative resolution of restriction in up-gaze after the repair of the linear floor fracture. The urgency for intervention should be based on the merits of each case. As a general guideline, trapdoor fractures need to be attended to within a week if possible, and open fractures, especially in older children, can be managed similar to the adult fractures. A randomized controlled study of the outcomes of linear blow-out fractures managed at different durations after initial insult from injury could provide better recommendations on the timing of intervention. Our study showed that despite the variation in the time from injury to surgical intervention, pediatric orbital blow-out fractures have good long-term (up to 9 months) outcome as long as the intervention was not delayed for too long. All ophthalmologists should always review the scans themselves and correlate with clinical findings to determine the diagnosis and appropriate course of management. REFERENCES 1. Bansagi ZC, Meyer DR. Internal orbital fractures in the pediatric age group: characterization and management. Ophthalmology. 2000;107:829Y836. 2. de Man K, Wijngaarde R, Hes J, et al. Influence of age on the management of blow-out fractures of the orbital floor. Int J Oral Maxillofac Surg. 1991;20:330Y336. 3. Beirne JC, Dhaif G. Orbital floor fractures in young children. J Craniomaxillofac Surg. 1995;23:401. 4. Sires BS, Stanley RB Jr, Levine LM. Oculocardiac reflex caused by orbital floor trapdoor fracture: an indication for urgent repair. Arch Ophthalmol. 1998;116:955Y956. 5. Levin LM, Kademani D. Clinical considerations in the management of orbital blow-out fractures. Compend Contin Educ Dent. 1997;18:593, 596Y598, 600. 6. Wilkins RB, Havins WE. Current treatment of blow-out fractures. Ophthalmology. 1982;89:464Y466. 7. Jordan DR, Allen LH, White J, et al. Intervention within days for some orbital floor fractures: the white-eyed blowout. Ophthal Plast Reconstr Surg. 1998;14:379Y390. 8. Gerbino G, Roccia F, Bianchi FA, et al. Surgical management of orbital trapdoor fracture in a pediatric population. J Oral Maxillofac Surg. 2010;68:1310Y1316. 9. Grant JH, Patrinely JR, Weiss AH, et al. Trapdoor fracture of the orbit in a pediatric population. Plast Reconstr Surg. 2002;109:482Y489. 10. Criden MR, Ellis FJ. Linear nondisplaced orbital fractures with muscle entrapment. J AAPOS. 2007;11:142Y147. 11. Lane K, Penne RB, Bilyk JR. Evaluation and management of pediatric orbital fractures in a primary care setting. Orbit. 2007;26:183Y191.
www.apjo.org
Copyright © 2012 by Asia Pacific Academy of Ophthalmology. Unauthorized reproduction of this article is prohibited.
269
Pediatric Orbital Blow-Out Fractures: Surgical Outcomes Jamie Y. Ng, MRCS Ed,* Sundar Gangadhara, FRCS,* Wong Wanling, PhD,Þ and Shantha Amrith, FRCOphth*
Purpose: The objective of this study was to evaluate the features, timing of intervention, complications, and outcomes of patients who underwent surgery for pediatric orbital blow-out fractures. Design: This was a retrospective case review studying all cases of pediatric orbital blow-out fractures that underwent surgical intervention from 2000 to 2009 in a tertiary ophthalmic center in Singapore. Methods: Case notes review of all cases of pediatric orbital blow-out fractures repaired surgically was carried out for demographics, time to intervention, and outcome. How the various factors may affect the outcome was statistically analyzed. Results: A total of 23 patients were reviewed. There were 21 patients (91.3%) with floor fractures, 1 (4.3%) with medial wall fracture, and 1 (4.3%) with combined medial wall and floor fracture. Of the 21 cases of floor fracture, 20 cases had radiological images available for review, 17 (85%) were linear, and 3 (15%) were open. Patients with open fractures were older with a mean age of 14.0 years as compared with 12.6 years in those with linear fractures. The mean duration from injury to intervention was 13.0 days. Preoperatively, 14 (60.9%) had diplopia, but only 4 cases (17.4%) had diplopia in extreme gaze at the last follow-up after surgery (18 months). Outcome was not affected by age, duration between injury and intervention, type of fracture, and nature of implants used. Conclusions: Our study demonstrated that pediatric orbital blow-out fractures had good outcomes. It might also suggest the possibility of not having to manage all linear fractures as a surgical emergency. Key Words: blow-out fracture, white eye, pediatric (Asia-Pac J Ophthalmol 2012;1: 265Y269)
T
he management of orbital blow-out fractures has always been a topic of controversy, and pediatric fractures are certainly not as well studied as adult fractures. The greater elasticity in pediatric bones as compared with adults results in greenstick deformity that manifests as trapdoor fractures.1Y7 Soft tissues such as extraocular muscle, intermuscular septa, and fat may become entrapped, resulting in significant restriction in eye movement. This might be without suggestive periocular signs of trauma and hence known as ‘‘white-eyed blow-out fracture’’ (Fig. 1), which is often missed.8 The purpose of our study was to evaluate the clinical, radiological features, and timing of intervention and correlate them with the surgical outcomes of pediatric orbital blow-out fracture repair.
that underwent surgical intervention by 2 senior oculoplastic surgeons from January 2000 to December 2009 was conducted. The age of 16 years was chosen as the cutoff age for inclusion into the study based on our institutional guidelines. There were a total of 23 patients. All patients underwent computed tomography (CT) scans. Axial, coronal, and in some cases sagittal reconstructions in both bone and soft tissue windows were read both by a radiologist and as well as the oculoplastic surgeon. Clinical evaluation included a detailed history and complete ophthalmologic examination including mechanism of injury, symptoms of nausea and vomiting, associated ocular injuries, diplopia in primary as well as different positions of gaze, extraocular eye movements, enophthalmos, and infraorbital paresthesia. Duration from injury to ophthalmic referral and duration from injury to intervention were also noted. Surgical factors such as the approach to the orbit, type of implant (according to patient’s and parent’s choice), and complications were examined. The outcome of surgery was measured by noting the improvement in diplopia and extraocular eye movements, resultant enophthalmos, and any residual complications. All patients received prophylactic systemic antibiotics (amoxicillin and clavulanate) postoperatively. Linear fractures (trapdoor), as the name suggests, are defined as orbital floor fractures that extend anteroposteriorly, allowing the herniation of orbital contents and entrapping them. Open fractures are usually hinged medially and have a gaping defect that allows the orbital contents to herniate through. Minimal diplopia was defined as diplopia present only in extreme gaze, moderate as diplopia present from 15 to 25 degrees from central fixation, and severe when diplopia was present at within 15 degrees from central fixation. The extraocular eye movements were likewise graded.
RESULTS All surgeries were performed by either of 2 senior oculoplastic surgeons in our center, and all surgeries were performed through a transconjunctival route, most of them not requiring cantholysis. The indications for surgery included clinical or radiological evidence of entrapment of tissue, large fracture greater than 50% of orbital floor, diplopia (any moderate or severe diplopia and/ or restriction of eye movements in Q1 positions of gaze), and persistent and significant enophthalmos greater than 2 mm.
MATERIALS AND METHODS With the approval of the institutional review board, a retrospective case review of all pediatric orbital blow-out fractures From the *Department of Ophthalmology, National University Health System, Singapore; †Singapore Eye Research Institute, Singapore. Received for publication April 16, 2012; accepted May 15, 2012. The authors have no conflicts of interest to declare. Reprints: Jamie Y. Ng, MRCS Ed, Department of Ophthalmology, National University Health System, 1E, Kent Ridge Rd, Tower Block, Level 7, Singapore 119228. E-mail: [email protected]. Copyright * 2012 by Asia Pacific Academy of Ophthalmology ISSN: 2162-0989 DOI: 10.1097/APO.0b013e31825f 8976
Asia-Pacific Journal of Ophthalmology
&
FIGURE 1. A child with a right white-eyed blow-out fracture of the orbit with severe limitation in up-gaze.
Volume 1, Number 5, September/October 2012
www.apjo.org
Copyright © 2012 by Asia Pacific Academy of Ophthalmology. Unauthorized reproduction of this article is prohibited.
265
Asia-Pacific Journal of Ophthalmology
Ng et al
Patient demographics are shown in Table 1. The mean age was 12.9 (SD, 2.6) years (range, 5Y16 years). Sixteen had linear fractures, and 4 had open fractures. The group with open fractures was slightly older than the group with linear fractures (mean, 14.0 vs 12.6 years). This was not statistically significant. The patients were predominantly male (n = 21; 91.3%). Of the 23 children operated on, the majority were Chinese (n = 14; 60.9%) followed by 7 Malays (30.4%) and 2 Indians (8.7%). There was thus a slight overrepresentation of Malays compared with the racial profile of Singapore, reported to consist of 74.2% Chinese, 13.4% Malays, 9.2% Indians, and 3.2% others. The mean duration of postoperative follow-up was 9.6 (SD, 8.0) months (range, 1Y42 months; median, 8.0 months); 3 cases defaulted follow-up after 1 month. Vasovagal symptoms were
TABLE 1. Demographics (N = 23) Demographics Age, mean (SD), y Linear factures Open fractures Sex, n (%) Male Female Race Chinese Malay Indian Follow-up, mean (SD), mo Vasovagal symptoms, n (%) Yes No Time from injury to ophthalmic referral, n (%) Same day 1 d postinjury 2 d postinjury 3 d postinjury 4 d postinjury 5 d postinjury Associated injuries, n (%) Yes No Laterality, n (%) Right Left CT findings,* n (%) Floor fracture Medial wall Combined Type of floor fracture* Linear Open Nature of implants (n = 21), n (%) Medpor Bioresorbable Supramid
Range 12.9 (2.6) 12.6 (2.9) 14.0 (1.1) 21 (91.3) 2 (8.7) 14 (60.9) 7 (30.4) 2 (8.7) 9.6 (8.0) 5 (22.7) 17 (77.3)
12 (52.1) 5 (21.7) 3 (13.0) 1 (4.4) 1 (4.4) 1 (4.4) 15 (65.2) 8 (34.8) 11 (47.8) 12 (52.2)
www.apjo.org
1.0Y42.0
Volume 1, Number 5, September/October 2012
TABLE 2. Mechanism of Injury (N = 23) n (%) Assault Sports Domestic accidents Pedestrians in road traffic accident
8 (34.8) 8 (34.8) 4 (17.4) 3 (13.0)
reported in 5 patients (22.7%). Majority (73.8%) was referred for ophthalmic evaluation within a day after injury, and the remaining within 5 days from injury. Hence, the mean time from injury to an ophthalmic referral was 1.0 day. At presentation, 15 patients (65.2%) had associated injuries. The most commonly associated injury was periorbital hematoma (n = 10; 41.7%); others included 3 upper eyelid lacerations (12.5%), 2 commotio retinae (8.3), and 1 case of microscopic hyphema (4.2%). Table 1 also describes the mode of injury. Assaults and sports injuries were equally common followed by domestic accidents and road traffic accidents. Sports activities included soccer (n = 4), baseball (n = 2), and softball (n = 2). Table 2 summarizes the fracture characteristics as seen in the CT, the time of intervention, and the outcome. The CT scans were reviewed by radiologists and ophthalmologists. For 1 case of floor fracture, the scans were done elsewhere and not available for review. Of the 20 cases that were isolated floor fractures, 17 (85.0%) were linear (Fig. 2), and 3 (15.0%) were open. Three (13.6%) of 22 scans were reported by radiologists as without any orbital fractures, but because of a strong clinical suspicion and subtle changes in CT scan suggestive of a fracture, surgery was performed, and the presence of fracture was confirmed in all 3. An example is shown in Figure 3. A detailed breakdown of duration from injury to surgery is given in Table 3. The median duration from injury to time of intervention was 13.0 (SD, 17.0) days (range, 0Y74 days; median, 10.0 days). Ten cases had surgery within 1 week of injury, and 6 others had surgery between 1 and 2 weeks after injury. The indication for surgery in the case that was operated 74 days after injury was residual enophthalmos. Mean time from injury to surgical intervention for linear fractures was 13.7 days and for open fractures was 14.8 days. Patients were evaluated for preoperative and postoperative infraorbital paresthesia, enophthalmos, and diplopia. Four patients (17.4%) had infraorbital anesthesia preoperatively, and it persisted in 2 cases beyond 1 month after surgery. Table 4 shows the comparison between preoperative and postoperative diplopia. Only 14 patients (60.9%) had preoperative diplopia, among whom 3 were present in primary position. None of
20 (90.9) 1 (4.5) 1 (4.5) 17 (85.0) 3 (20.0) 15 (71.4) 4 (19.0) 2 (9.5)
*Imaging films were not available for 1 subject.
266
5.0Y16.0 5.0Y16.0 12.0Y15.0
&
FIGURE 2. Computed tomographic imaging showing a linear right orbital floor fracture in the coronal cut with entrapment of orbital contents. * 2012 Asia-Pacific Journal of Ophthalmology
Copyright © 2012 by Asia Pacific Academy of Ophthalmology. Unauthorized reproduction of this article is prohibited.
Asia-Pacific Journal of Ophthalmology
&
Volume 1, Number 5, September/October 2012
FIGURE 3. Computed tomographic imaging showing a left linear orbital floor fracture that was initially missed by the radiologist.
the patients had poor vision limiting the assessment of diplopia. Overall, 4 patients (17.4%) had residual mild diplopia after surgery at the last follow-up (18 months), and in all cases diplopia was far
Pediatric Orbital Blow-Out Fractures
less symptomatic than at presentation. None had significant residual diplopia that affected their daily activities. Overall, the diplopia in primary position and up-gaze showed significant improvement (P = 0.006 and 0.014, respectively). Preoperative exophthalmometric readings were not included in the analysis as many patients had proptosis due to the orbital hematoma and edema, and a considerable number underwent surgery before it settled. Two patients (8.7%) had mild residual enophthalmos (3 mm) postoperatively, but both were asymptomatic (Table 5). Residual diplopia in any gaze position in relation to time from injury to intervention was analyzed. There were no significant differences in this final outcome in the early (G48 hours) and late (92 weeks) intervention groups (P = 0.876). Patients who presented with diplopia regardless of severity and positions of gaze were operated on at a mean duration of 7.2 (SD, 8.2) days (range, 0Y26 days), as compared with children without diplopia at presentation [mean, 22.1 (SD, 23.0) days; range, same day to 74 days]. Patients with any
TABLE 3. Fracture Characteristics, Time of Intervention, and Outcome Case Number 1* 2 3 4
Type of Fracture
Image (CT)
TrapdoorVjunction of medial wall and floor Trapdoor
A, C, S A, C, S
Trapdoor Open fracture of floor and medial wall Trapdoor Trapdoor
5 6 7
TrapdoorVjunction of medial and floor Open Trapdoor Trapdoor
8 9* 10 11 12 13 14
A, C, S A, C, S A, C, S A, C A, C A, C, S A, C A, C
15 16 17
Medial wall A, C, S Open A, C Trapdoor A, C TrapdoorVjunction of medial A, C, S wall and floor Scans not available for review Trapdoor A, C, S Trapdoor A, C
18*
Trapdoor
A, C
19 20
Trapdoor Trapdoor
A, C, S A, C, S
21 22 23
Open Trapdoor TrapdoorVjunction of medial wall and floor
A, C A, C A, C
Injury to Surgery Time
Outcome
Entrapment of fat
0
Good
Entrapment of fat and muscle sheath Entrapment of fat IR muscle inside fracture
0
Diplopia in extreme gaze
0 0
Diplopia in extreme gaze Good
Entrapment of IR muscle Entrapment of fat and muscle sheath Entrapment of IR muscle
1 1
Good Good
1
Diplopia in extreme gaze
IR muscle inside the fracture Entrapment of muscle sheath Entrapment of fat and muscle sheath MR inside the fracture IR muscle inside fracture Entrapment of fat Entrapment of fibrous connective tissue
1 4 5
Good Good Good
9 10 11 12
Good Mild enophthalmos Good Good
13 13 15
Good Mild enophthalmos Diplopia in extreme gaze
16
Good
20 26
Good Good
28 40 74
Good Good Good
Image Findings
Entrapment of fat Entrapment of fat and muscle sheath Entrapment of fat and muscle sheath Entrapment of IR muscle Entrapment of fat and muscle sheath IR muscle inside the fracture Entrapment of fat Entrapment of fibrous connective tissue
*Missed by radiologist. A indicates axial; C, coronal; S, sagittal; IR, inferior rectus; MR, medial rectus.
* 2012 Asia-Pacific Journal of Ophthalmology
www.apjo.org
Copyright © 2012 by Asia Pacific Academy of Ophthalmology. Unauthorized reproduction of this article is prohibited.
267
Asia-Pacific Journal of Ophthalmology
Ng et al
&
Volume 1, Number 5, September/October 2012
TABLE 4. Duration Parameters Range Duration from injury to surgery, no. cases (%) G48 h Q48 h to G1 wk Q1 wk to G2 wk Q2 wk Mean time from injury to surgical repair, mean (SD), d Overall Presented with initial diplopia that resolved With residual postoperative diplopia Linear fractures Open fractures
FIGURE 4. Preoperative photograph showing right eye limitation in up-gaze in a child with right linear floor fracture.
8 (34.8) 2 (8.7) 6 (26.1) 7 (30.4)
13.0 (17.0) 7.2 (8.2)
0Y74
4.0 (7.3) 13.7 (19.5) 14.8 (11.8)
residual diplopia after surgery were intervened at an average of 4.0 days (same day to 15 days) as compared with 7.2 days for those who presented with diplopia initially that resolved after surgery. There was no statistical difference in the outcome of these 2 groups of patients (P = 0.375). Postoperative diplopia was also not found to be significantly influenced by
TABLE 5. Comparison of Preoperative and Postoperative Diplopia and Enophthalmos Preoperative, Postoperative, n (%) n (%) Diplopia Yes No Primary position Yes No Up-gaze None Mild Moderate Severe Down-gaze None Mild Moderate Severe Horizontal None Mild Moderate Severe Enophthalmos at last visit after surgery† Yes No
DISCUSSION P* 0.006
14 (60.9) 9 (39.1)
4 (17.4) 19 (82.6)
3 (13.0) 20 (87.0)
0 (0.0) 23 (100.0)
11 (47.8) 6 (26.1) 2 (8.7) 4 (17.4)
20 (87.0) 3 (13.0) 0 (0.0) 0 (0.0)
20 (87.0) 2 (8.7) 0 (0.0) 1 (4.3)
23 (100.0) 0 (0.0) 0 (0.0) 0 (0.0)
20 (87.0) 1 (4.3) 1 (4.3) 1 (4.3)
22 (95.7) 1 (4.3) 0 (0.0) 0 (0.0)
0.006
0.014
0.233
0739
V
2 (8.7%) 21 (91.3%)
V
*P value from Fisher exact test. †Preoperative measurements not included as some patients had proptosis due to hematoma and edema.
268
www.apjo.org
age (P = 0.596), fracture type (P = 0.861), or implants used (P = 0.999). Of the 23 cases, 22 had implants, but the type of implant was unknown in 1 case. One patient did not require an implant, as there was no significant residual defect after the inferior rectus entrapment was released. Porous polypropylene (MedporBarrier Sheet; Porex Surgical Inc, College Park, Ga) (15; 71.4%) was the most commonly used implant, followed by bioresorbable PolyMax (Synthes, Oberdorf, Switzerland) (4; 19.0%). Supramid (Ethicon, Somerville, NJ) was used in 2 patients (9.5%) who underwent surgery before 2002. These findings are indicated in Table 1. We had only 1 case with early postoperative retrobulbar hemorrhage presenting with optic neuropathy hours after surgery in a patient with large medial wall fracture, which was repaired with Medpor implant. This was managed conservatively with intermittent cold compression and intravenous methylprednisolone, and the patient recovered well without residual visual deficit. Other complications from the surgery such as infection, vision loss, or implant-related complications were not seen in any case. Hence, the overall rate of complication was 4.3%.
Current recommendations for surgical intervention in orbital blow-out fractures include muscle and soft tissue entrapment resulting in significant diplopia, enophthalmos of more than 2 mm, and orbital defects involving greater than 50% of either the medial wall or the floor. The urgency of management of adult and pediatric cases has always differed. Surgery in the pediatric group has been conducted more urgently in view of the greater potential for entrapment resulting in muscle ischemia and fibrosis.1,2,4,9 It has also been demonstrated that even with early intervention, limitation in eye movements may be present in the early postoperative period because of ischemia, edema, and hemorrhage, but they tend to recover completely in weeks to months.10 Most studies in the literature on pediatric blow-out fractures are limited by the small numbers similar to our study, making statistical analysis impossible. The ischemia and necrosis of the trapped extraocular muscle (inferior rectus in most cases) has not been clearly demonstrated. Various studies suggest early intervention in pediatric population, but the conclusions derived about timing of surgical intervention versus results are anecdotal. A magic number for trauma to surgical intervention time may not exist in pediatric blow-out fractures because the urgency of intervention may be dependent on individual case’s clinical and CT scan findings. Our study suggests that there are no significant differences in the final outcome in the early (G48 hours) and late (92 weeks) intervention groups. This may indicate that the diplopia outcome may be dependent on the severity of the initial injury and fracture. Although a 1-week interval between injury and intervention is acceptable in most cases, one should always take the age of the patient, clinical signs, and the nature of entrapment into account while managing pediatric patients with blow-out fractures. It appears that our physicians in the frontline are able to recognize the clinical features of pediatric orbital fractures and * 2012 Asia-Pacific Journal of Ophthalmology
Copyright © 2012 by Asia Pacific Academy of Ophthalmology. Unauthorized reproduction of this article is prohibited.
Asia-Pacific Journal of Ophthalmology
&
Volume 1, Number 5, September/October 2012
FIGURE 5. Early postoperative photograph demonstrating full range of movement in up-gaze in the same child.
are able to act quickly as evidenced by referrals within a day to the oculoplastic team in more than three fourths (73.8%) of the patients. It will be useful to do a prospective data collection about the flow of such patients within the hospital to determine where the delay is occurring, so that the hospital processes can be modified to shorten the duration between trauma and surgery. Although the follow-up period in 3 of our cases was only 1 month, all of them had no diplopia or enophthalmos at their last visit, and therefore, our statistical analysis was not adversely affected by the short follow-up. Computed tomographic scans have been widely accepted and useful in the diagnosis of orbital fractures. Orbital fractures have been reportedly missed radiologically on CT scan,1,7 and this was evident in our study as well. White-eyed blow-out fractures have subtle signs both physically and radiologically. Hence, all scans should be carefully evaluated and correlated to clinical findings before ruling out the diagnosis of a fracture. Children with trapdoor fractures tend to have nausea and vomiting.1,4,11 It has been related to the oculocardiac reflex, and such symptoms should alert all ophthalmologists, especially in young children who may not be able to express symptoms of diplopia. Our study reports a rate of vasovagal symptoms of 22.7%, which is significantly lower than 75% as reported by Lane et al.11 Our study, being retrospective, may not have captured such systemic symptoms because of improper documentation in the case records. Although no implant-related complications were seen in this series, bioresorbable implants are probably better in pediatric orbital blow-out fractures. It is ideal for linear fractures where the defect tends to be small and a long-term structural support is not required. The parents also like the idea of their children not having to live with permanent implants in their bodies. However, in large fractures, it is still not clear if the bioresorbable implants afford the kind of structural support necessary after complete resorption; therefore, a more permanent implant such as Medpor or titanium may be necessary to support tissues. We acknowledge that our study results are limited by the small sample size. The other limitation is that we chose to study diplopia as the outcome measure rather than restriction of eye movement. Although the decision to intervene surgically was based on motility restriction combined with forced duction test and CT scan findings, not all children complained of diplopia. It was difficult to quantify the motility restriction because of noncooperation from children partly because of the pain associated with the eye movements.
* 2012 Asia-Pacific Journal of Ophthalmology
Pediatric Orbital Blow-Out Fractures
Our study confirms that pediatric blow-out fractures have good outcome without significant complications after surgical intervention and suggests that not all cases need to be intervened as an emergent condition. Figures 4 and 5 demonstrate an example of good postoperative resolution of restriction in up-gaze after the repair of the linear floor fracture. The urgency for intervention should be based on the merits of each case. As a general guideline, trapdoor fractures need to be attended to within a week if possible, and open fractures, especially in older children, can be managed similar to the adult fractures. A randomized controlled study of the outcomes of linear blow-out fractures managed at different durations after initial insult from injury could provide better recommendations on the timing of intervention. Our study showed that despite the variation in the time from injury to surgical intervention, pediatric orbital blow-out fractures have good long-term (up to 9 months) outcome as long as the intervention was not delayed for too long. All ophthalmologists should always review the scans themselves and correlate with clinical findings to determine the diagnosis and appropriate course of management. REFERENCES 1. Bansagi ZC, Meyer DR. Internal orbital fractures in the pediatric age group: characterization and management. Ophthalmology. 2000;107:829Y836. 2. de Man K, Wijngaarde R, Hes J, et al. Influence of age on the management of blow-out fractures of the orbital floor. Int J Oral Maxillofac Surg. 1991;20:330Y336. 3. Beirne JC, Dhaif G. Orbital floor fractures in young children. J Craniomaxillofac Surg. 1995;23:401. 4. Sires BS, Stanley RB Jr, Levine LM. Oculocardiac reflex caused by orbital floor trapdoor fracture: an indication for urgent repair. Arch Ophthalmol. 1998;116:955Y956. 5. Levin LM, Kademani D. Clinical considerations in the management of orbital blow-out fractures. Compend Contin Educ Dent. 1997;18:593, 596Y598, 600. 6. Wilkins RB, Havins WE. Current treatment of blow-out fractures. Ophthalmology. 1982;89:464Y466. 7. Jordan DR, Allen LH, White J, et al. Intervention within days for some orbital floor fractures: the white-eyed blowout. Ophthal Plast Reconstr Surg. 1998;14:379Y390. 8. Gerbino G, Roccia F, Bianchi FA, et al. Surgical management of orbital trapdoor fracture in a pediatric population. J Oral Maxillofac Surg. 2010;68:1310Y1316. 9. Grant JH, Patrinely JR, Weiss AH, et al. Trapdoor fracture of the orbit in a pediatric population. Plast Reconstr Surg. 2002;109:482Y489. 10. Criden MR, Ellis FJ. Linear nondisplaced orbital fractures with muscle entrapment. J AAPOS. 2007;11:142Y147. 11. Lane K, Penne RB, Bilyk JR. Evaluation and management of pediatric orbital fractures in a primary care setting. Orbit. 2007;26:183Y191.
www.apjo.org
Copyright © 2012 by Asia Pacific Academy of Ophthalmology. Unauthorized reproduction of this article is prohibited.
269
