Abstract
Introduction
Complex orbital fractures require multi-specialty input. The United Kingdom currently has no agreed best practice guidelines. A systematic review and meta-analysis found a significantly increased adverse outcome of persistent diplopia in patients operated on more than 14 days after initial injury. At our tertiary centre, we felt patient outcomes may be affected by investigative and referral pathway limitations so evaluated to assess and attempt to improve these.
Methods
This was a two-stage service evaluation study of all orbital trauma patients requiring specialist oculoplastic and orbital input. The initial phase included all orbital trauma patients referred to our department in April–November 2014. We assessed relevant key performance indicators and then implemented changes to our referral system, educated referring departments and formed a multidisciplinary team clinic to manage complex trauma. We then re-evaluated, collecting data on all patients referred March–May 2015.
Results
Most patients were male and victims of assault. Following intervention, there was a significant reduction in the injury to eye clinic review time (3.45 weeks to 2.22 weeks, p = 0.01) and multidisciplinary team review time (3.86 weeks to 2.71 weeks, p = 0.04). Injury to discharge time was significantly reduced from 6.15 weeks to 3.36 weeks (p = 0.002).
Discussion
Our study shows that a multidisciplinary team approach and structured pathways can improve assessment and intervention times for orbital trauma patients. This may have a beneficial effect on patient outcomes. In the absence of guidelines in the United Kingdom, we believe similar improvements could be implemented by other national centres and outcomes evaluated to assist future guidance formation.
Introduction
The orbit is a common site affected by trauma. Fractures of the orbital rim make up around 20–25% of all facial fractures. 1 These are commonly as a result of assault or accidental injury. 1 The management of orbital trauma is multidisciplinary and involves a number of specialties. Repairs of orbital fractures may be performed by ophthalmologists, ear, nose and throat (ENT), oral and maxillofacial surgeons (OMFS) or plastic surgeons depending on the local department expertise and availability. Radiology input is required for almost all cases. The involvement of an ophthalmologist with an oculoplastic specialist interest is important as around one third of patients with an orbital rim fracture on imaging will have associated ocular pathology including reduced visual acuity (VA) or restricted eye movements. 2 There is also a high rate of ocular injury in asymptomatic patients with rates between 26% and 40%.3,4
Within the sphere of orbital trauma, there are relatively few guidelines specifically addressing on the assessment and management of patients following trauma to the orbit. The College of Optometrists in the UK provide guidance to members on blunt trauma to the eye 5 but this does not specifically include management of orbital injures and has limited relevance to the hospital clinician. In the UK, there are no current local, regional or national guidelines on the indications for ophthalmology review of patients with orbital injuries and the recommended examinations and investigations.6,7 The authors are not aware of published evidence-based guidelines in other countries covering these issues.
There is no consensus internationally on who should perform repairs to the orbit in order to achieve optimal outcomes and we are not aware of published research on what the optimum pathway for management of such patients or whom should comprise the team managing them. 8
What has been more recently established is the time between injury and when surgical repair should be conducted, should this be required. A recent systematic review 9 of five retrospective chart review studies which assessed outcomes following fracture repair within or longer than 14 days from injury had shown evidence of improved outcomes for orbital blowout fractures repaired within 14 days of injury, particularly with regard to post-operative persistent diplopia. This highlights the importance of streamlining a pathway that required multiple inputs from different specialties to optimise outcomes. This review post-dated our study, however in our own unit, we were aware of there were multiple potential points of delay introduced at each juncture of the patient pathway at different units and no uniform method of information transfer to facilitate patient care. We felt that, as a result of this, a number of patients were left with persistent diplopia.
Our aim was to identify the delays in the patient pathway and assess the proportion of our patients who had complications. We also took this opportunity to use the data collected to obtain a better understanding of the demographics of users of our service and the cause of the injuries they present with. We hope that we can use the knowledge gained to inform future guidance on the ophthalmic assessment of patients following orbital trauma.
Methods
Study design
This is a retrospective, quasi-experimental case series of patients seen in an ophthalmology clinic at a regional facial trauma centre following orbital trauma.
Setting
Our study was carried out at a state-run academic health trust in North West London, UK, with three principal sites and around 1100 patient beds. It receives around 80,000 admissions per annum. 10 The trust has a full range of medical and surgical departments and in the case of facial trauma, it is the tertiary referral centre for patients in North West London and environs, receiving referrals from surrounding health trusts and covering a population of approximately four million people. The trust has a full range of imaging capabilities for trauma patients, a 24 hour on call OMFS team, dedicated inpatient facilities for OMFS patients and weekly ophthalmic and OMFS outpatient clinics and theatre lists for orbital trauma management including one joint theatre list.
Population
The population for our study was comprised of all patients who had received trauma to the orbit and required OMFS and ophthalmology review. Indications for referral for ophthalmology review included: where a decision for surgery has been made but there was concern that there is concomitant ocular and/or ocular motility injury that required assessment and documentation preoperatively and to ensure that open reduction and internal fixation (ORIF) surgery will not jeopardise any eye injury; where the decision for surgery was borderline so a second opinion from the oculoplastic surgeon was sought or that there had been an unexpected outcome following surgery and an opinion was sought prior to re-operation.
As this study is a pilot study, power calculations were not performed prior to the study commencing as expected differences between the groups could not accurately be estimated.
Data collection
We undertook a retrospective study of consecutive patients referred to the Oculoplastic service from OMFS between April and November 2014. Data collection included patient’s sex, age, injury date and mechanism of injury. Dates of hospital attendances, ophthalmic injuries and procedures, and surgical procedures and complications, particularly persistent diplopia and visual impairment were collected retrospectively from electronic health records and paper records where available. The patient’s images and radiology reports were accessed using the radiology PACS imaging system to assess for the presence of fractures and/or ocular injuries. Patients were excluded from the study if they sustained trauma prior to three months before the date of referral or if insufficient data were available from hospital records. A database was created to facilitate data collection as shown in Appendix 1. All data were collected by MG and NM in early 2015. Records were reviewed up until the date of discharge or failure to attend scheduled reviews in the OMFS and/or ophthalmology service.
Following the interventions (below), the same data were collected for the second section of our study using identical methods, by MG, on patients referred from OMFS to ophthalmology between March and May 2015. Data were collated in a new version of the database in Appendix 1. Data collection for the second section of the study was terminated early as two of the authors changed employment in August 2015 and so insufficient follow-up data were available to them for patients referred after May 2015.
Intervention
Following the initial study period, a new orbital trauma lead was appointed to OMFS (MP) and both the senior authors (VL and MP) identified a series of key performance indicators that needed improvement. This was to both to reduce the resource required for management of these patients and identify those patients requiring surgery early to improve waiting times and potentially improve outcomes.
8
These key performance indicators (KPIs) included:
Percentage of patients seen and assessed in accident and emergency, radiology, oral and maxillofacial surgery and ophthalmology within two weeks following their injury. Target 80% Percentage of patients undergoing CT scan as their primary imaging modality and this being performed prior to clinic assessment. Target 100% Percentage of patients undergoing definitive surgery if needed within 2 weeks of injury. Target 100% Percentage of patients not attending booked appointments. Target 0% Percentage of patients being referred to eye casualty units without a documented indication to do so. Target 0%
We presented our initial study findings and recommendations to at both the ophthalmology and OMFS tertiary referral unit clinical governance meetings. Information regarding the results and recommendations was disseminated via e-mail to clinical leads of Accident and Emergency (A + E) departments in the population area. We introduced a new proforma to expedite referrals between the two departments as well as setting up a trauma email group to ease rapid referral and communication on the secure NHS electronic mail system (nhs.net). All orbital trauma cases were discussed with MP as to whether an urgent oculoplastic opinion was needed. A dedicated administrator was designated to handle the referrals from OMFS to Oculoplastics in a timely fashion allowing most urgent patients to be seen and operated on in the same week after being assessed in both services. Additionally, prior to our repeat service evaluation, and a monthly orbital multidisciplinary team (MDT) clinic was initiated to facilitate and optimise the management of complex post traumatic surgical cases. Figure 1 indicates a visual illustration of our pathway changes.

Map of all sites from which referrals are received. Solid circles: referring units; dashed circles: referral centres. Map © Google Inc.
Data analysis
The primary aim of our study was to determine whether the above interventions produced a significant increase in the number of patients undergoing surgery for orbital trauma where indicated within two weeks of injury.
Secondary aims included assessing whether our interventions produced a significant reduction in the time for patients to be assessed by OMFS, radiology, A + E and ophthalmology, whether there was a significant reduction in the time from injury to surgery and also to obtain data on the demographics and mechanism of injuries sustained by our patients.
Statistical analysis was conducted using Microsoft Excel (Microsoft Inc., Seattle, WA) and MedCalc for Windows (MedCalc Software, Ostend, Belgium). Averages for values of time are expressed as medians. Chi squared testing was used to compare data between groups and unpaired t-test and Mann–Whitney U tests were used to assess the difference between normal and non-normally distributed data respectively. P values of ≤0.05 were taken to be statistically significant.
Ethical considerations
This study was approved by the Research and Development Department of the trust in which it was conducted. As it was purely retrospective and used anonymised data, it was determined formal ethical approval and individual patient consent was not required. The study adhered to the tenants of the Declaration of Helsinki. No financial support was obtained for this study.
Results
Initial evaluation: April to November 2014
Seventy-eight eyes of 78 patients met the inclusion criteria for the initial evaluation and were included. Their demographics and presenting features can be seen in Table 1. The majority of patients were male with a mean age of 41 years. Looking more closely at patient demographics, there were two peak incidences in age groups, patients in between 20 and 40 due to assaults and others over 65 due to falls.
Summary of patient parameters and evaluation timescales between the original study and the post intervention study (*=statistically significant).
Out of 61 patients 12 (20%) were reviewed by OMFS, Radiology, A + E and Ophthalmology within two weeks of injury. Patients were seen at a median time of 3.45 weeks from injury in the eye clinic and it took a median of 3.86 weeks for patients to be seen by the full MDT following their injury.
Imaging was available for 72 (92%) of the patients and 64 (89%) of these patients had undergone CT scanning.
Surgery was carried out within two weeks of injury for 3 patients out of the 22 who went on to have surgery (14%). Those having surgery had this performed at a median of 3.86 weeks after injury.
Of those referred, 17 (22%) did not attend their ophthalmology appointment despite all patients receiving communication of their appointment date and time via letter and text message. Seventeen (22%) patients were documented to have attended a walk-in eye emergency service prior to their review in our clinic but for 49 (63%) referrals, it could not be determined from the available records at our centre whether they had attended an eye casualty or not. Indications for eye casualty review were not available in most cases and so evaluation could not be performed.
Looking in more detail at the demographics and injuries of our patients, the most common mechanism of injury for all patients referred was assaults followed by falls (Figure 2). Of those patients who attended for review, 54 (89%) had a fracture of one or more orbital bones.

Illustration of the mechanism of orbital injury for all patients. Percentages may not add up to 100% due to rounding.

(a) Patient pathways prior to implementation of new trauma pathway. (b) Patient pathway following implementation of new pathway.
Thirteen (21%) patients seen were found to have or were previously documented as having an injury to the globe itself. This ranged from a mild traumatic uveitis to retinal tears or penetrating globe injuries. Of these, 4 (31%) patients required an ophthalmic surgical or laser procedure.
In this audit, 22 (36%) patients underwent a surgical repair of their fracture. Three (14%) patients had required revision surgery one for facial asymmetry and two for plate migration.
Evaluation following intervention: March to May 2015
Following implementation of the above changes, we re-audited 32 eyes of 32 patients referred between March and May 2015. A breakdown of the re-audit patient’s findings is found in Table 1 and Figure 2.
There was no significant difference in the proportion of male patients, patients not attending their appointments, patients having an ocular injury, patients with an orbital rim fracture or patients requiring ocular surgery. There was however a significant difference in the mean patient age in this second group, 31 years vs. 41 years (p = 0.01, 95% CI −17.65 to −2.35). This is likely to relate to a slightly lower number of patients over the age of 60 in the second group as opposed to the initial group (19% vs. 13%).
As a result of changes in our referral pathway, 38% of patients were seen in A + E, OMFS, radiology and ophthalmology within two weeks vs. 20% in our initial study. This approached but did not reach statistical significance (p = 0.08, 95% CI −1.82 to 38.71). There was however a significant reduction in the time between injury and review in the eye clinic, a median of 2.22 weeks (range 0.43–8 weeks) vs. 3.45 weeks (p = 0.01, Mann–Whitney U Test) and in the time between injury and review by the whole MDT a median of 2.71 weeks (range 1.14–8 weeks) vs. 3.86 weeks (p = 0.04, Mann–Whitney U Test). We feel that the reduction in time taken, particularly for ophthalmology reviews are clinically significant in allowing patients to have surgery in a more timely fashion to optimise outcome.
Imaging was available for 100% of patients at the time of clinic review. CT scanning was performed in 90% of patients which was not significantly different from the proportion having CT scans as the first definitive imaging in the initial audit (p = 0.88, 95% CI −15.21 to 12.33). This approaches our KPI target for this measure.
Surgery was performed within two weeks of injury for 2 out of the 11 patients who went on to have surgery (18%). This is not significantly different from that of the first group (p = 0.762, 95% CI −19.62 to 34.87). The median time from injury to surgery reduced from 3.86 weeks to 2.86 weeks which approached but did not reach statistical significance (p = 0.07) but the number of patients in these groups is small. We feel that this is a clinically significant reduction however as it reduces the average time to surgery by around 25% and brings patients to within six days of the optimum surgery time. Looking at all patients, the time from injury to definitive management or discharge was significantly reduced from 6.15 to 3.36 weeks (p = 0.002, Mann–Whitney U Test).
In the initial audit group, three patients (14%) required revision surgery whereas no patients to date from the re-audit group have undergone revision surgery in our unit or to our knowledge.
Despite our interventions to reduce DNA rates, 19% of patients did not attend their booked eye clinic appointment. This was not significantly different to the figure in the initial part of our study (p = 0.73, 95% CI −15.33 to 17.43). Only two patients (6%) were documented as having attended eye casualty, data for the remaining patients were not available so we were unable to evaluate our final KPI.
Discussion
Our study provides evidence that making changes to the pathways between the members of the MDT managing orbital trauma may have an effect on KPIs in the management of orbital trauma. There were significant reductions in the median time between injury, multi-disciplinary review and their management and discharge from the clinic. This is still some way from our target of 80%, implying further work is required. We have also shown a reduction in the time between patients having an orbital injury to having this surgically repaired when necessary and a small increase in the percentage of patients having surgery within two weeks of their injury following our interventions. This is still significantly below our target of 100% and so we continue to develop ways of improving this. It is likely to be difficult to reach our high target for this KPI due to delayed presentation of some patients following injury. When patients had surgery more quickly, there was a reduction in the number of patients requiring re-operation however due to the small sample size, this was not statistically significant.
A high number of CT scans were performed as a patient’s primary imaging modality of choice as in line with best practice documented elsewhere11,12 and this approached our target of 100%. Unfortunately, we were unable to accurately assess our final KPI on eye casualty attendance for patients following trauma and are planning another study to separately evaluate this.
We have additionally published a snapshot of the demographics and aetiologies behind patients attending a regional referral centre for orbital trauma in the United Kingdom, serving an urban population of approximately four million people.
We feel that the results of our study, whilst undertaken in the UK and concerning orbital trauma are relevant to an international audience and principles used are relevant to other patient pathways. Interventions such as modernising internal referral pathways and creating dedicated clinics for the management of complex patients between multiple specialities and disciplines is of relevance to all medical specialties. One example is where in patients with enterocutaneous fistulas, regular MDT meetings and a revision of the patient pathway produced a reduction in inpatient length of stay. 13 In hip and rib fractures, use of an MDT pathway has been shown to improve outcomes.14,15 In many disciplines, optimal management outcomes are achieved with a certain time period from illness or insult and our study suggests ways to reduce the time taken from presentation to management of injury or illness.
The general consensus among clinicians treating orbital trauma that the best window for surgical intervention for most orbital bony injuries (with the exception of orbital compartment syndromes such as white eye blowout 16 ) is within two weeks of the injury to allow the initial swelling to subside but before malunioned bone is likely to occur-requiring osteotomy.9,17,18 Surgery for orbital fractures performed more than eight days after injury has also shown in one study to lead to reduced post-operative motility compared to those operated more than eight days after injury. 19 The patient pathway involves many potential delays before management occurs at the regional centre, therefore it is important to expedite the patient journey so that the optimal window for surgery is not missed increasing the risk of long term adverse sequelae such as reduced motility and reduced field of binocular single vision.19,20
The population assessed in our study is comparable to that included in other studies. The preponderance of young males receiving orbital trauma as a result of assault is confirmed in other studies undertaken in other countries.1,21 Our study did however show a lower rate of facial trauma due to motor vehicle collisions than some other studies 22 perhaps reflecting a lower prevalence of drivers within our population area than more rural areas. In our two data sets, ocular injury following orbital trauma was 8% in our initial audit and 21% in the re-audit, making an overall rate of 17.5%, comparable to previous studies with a rate of ocular injury of between 10% and 27.6%.23,24
Patients with orbital trauma (particularly major trauma involving the orbit), should be assessed initially in the emergency department or equivalent location as the priority is assessment and optimisation of the airway, breathing and circulation.25,26 They then need to be assessed by the maxillofacial surgery team (or other surgical team with equivalent experience and skills in the management of orbital trauma) and in the case of suspected or actual ocular or orbital injury, an ophthalmologist specialising in this area. 17 As plain X- rays have limited role in the assessment of complex orbital trauma, 17 patients should have a fine slice CT of the orbits should be the primary imaging modality16,17 and so radiology must be part of the MDT.
Our study has a number of limitations. The study has a relatively small sample size. Reasons for this include the fact that our study only looked at patients with orbital trauma in whom OMFS clinicians felt that an ophthalmology review was warranted. Many patients would have been seen following orbital trauma without any symptoms or concerns of ocular injuries or motility restriction. There are a number of other eye units both specialising in and seeing patients with trauma and so these are not included in our sample. The two groups are not equal in number due to limitations in data collection in the latter arm and this also meant the duration of data collection was unequal in each group. As a retrospective study, some data were missing or could not be collected which may affect the results. The researchers were involved in the management of these patients (although those collecting the data were not) and so the researchers were not blinded introducing possible bias. As this study was an exploratory one, we were unable to carry out meaningful power calculations and so cannot conclude that our study had power to determine the exact effect of specific interventions. In future, for a full study, we would aim to use the experience gained from this study to re-evaluate our KPIs in a prospective manner with equal group sizes.
Another limitation is the failure to evaluate our last KPI on eye casualty attendance. Even in those patients that were seen by us, we had very little data from patients presenting acutely to other eye units following their injury before attending our unit. London has several eye units where patients can present as a walk in or be directed to. 27 Patients may have had an ocular injury diagnosed or treated there which was not included in our records. Across our entire study group of 110 patients, we were aware of 21 patients attending an eye casualty prior to their appointment; 3 of these did not attend their scheduled ophthalmology follow up, and 11 patients were positively documented as not having previously attended an eye casualty appointment. This means that for 77 patients (70%) no data were available on attendance at other eye units. We plan to address this separately as above.
In London, attendance at an eye casualty, particularly out of working hours, does not allow for a same day assessment by an oculoplastic surgeon and full orthoptic assessment. These should be performed in patients with or suspected to have an injury to the eye structures. 17 In our experience, inappropriate referral may lead to avoidable duplication of testing and resources such as orthoptic time and may confuse the patient about the need to attend the oculoplastic clinic when it this recommended by OMFS. Only seven of these patients attending eye casualty were found to have an ocular injury with four of these requiring an ophthalmic procedure as a result. Whilst other patients may have been reporting symptoms such as new flashing lights which would make urgent review appropriate, 28 up to two-thirds of these visits are potentially avoidable.
We have developed a guideline (see Table 2) to referring general casualty departments to determine which patients should be assessed within 24 hours of presentation and which are likely to be safe to be reviewed by ophthalmology as needed within a few days. This is based on symptoms likely to be suggestive of ophthalmic pathology requiring urgent intervention versus those which are commonly present in an orbital injury without ophthalmic injury. As the rate of ocular injury requiring ophthalmic intervention is relatively low in patients with orbital trauma,23,24 not all patients will require a review by eye casualty. A clinician’s judgement has the potential to reduce unnecessary visits to eye casualty whilst hopefully avoiding missed serious but reversible ocular pathology. From our experience, patients who are asymptomatic, are not undergoing surgical fracture repair and have normal visual acuities, eye movements and pupillary reactions are unlikely to benefit from an urgent ophthalmology review.
Indications for an ophthalmology review <24 hours from injury.
RAPD: relative afferent pupillary defect; SCH: subconjunctival haemorrhage.
Non-attendance is an important issue and despite sustained efforts to encourage patients to attend, around 1 in 5 patients failed to attend their ophthalmology appointment. This also limited the number of patients on whom we could obtain data on their patient journey. Our department uses telephone calls and text messaging to avoid any potential for problems with receiving letters in time for appointments and because our patient demographic is very familiar with this as a method of communication. In England, the overall DNA rate for outpatient appointments in the second quarter of 2016 was 9.9%. 32 Text messaging reminders have been shown to be effective at reducing DNA rates. 33 Studies previously have also shown that e-mail reminders can reduce DNA rates 34 and this is an additional measure which could be investigated in our unit.
In this study, we have demonstrated a lower rate of re-operation (19% vs. 0%) when the mean time to surgery was reduced from 3.86 to 2.86 weeks. This provides some evidence that reducing time to surgery may decrease need for re-operation However, it is unwise to draw any conclusion as to the optimum timing of orbital fracture surgery from this study as the sample size is small, there are confounding factors, a large number of our patients did not have a clinical indication for surgery and this study was neither designed nor powered to detect such a difference. We hope however that with continued work we will be able to reduce our time to surgery further and then can then look at outcomes with a view to providing further evidence of the benefit of early intervention.
From our data collected so far and our experiences, we would like to propose the following:
Recommendations:
A streamlined multi-disciplinary pathway between all those involved in the care of orbital trauma patients should be used. This has been shown to benefit outcomes in other forms of trauma14,15 Imaging should be performed using CT scanning11,12,25 and images and reports should be available to all clinicians involved in decisions regarding surgical intervention at the time of their review Patients should only be sent to eye casualty departments/clinics with a valid indication as outlined in Table 2. Any relevant information from such visits should be communicated to the orbital trauma unit Reviews by OMFS and ophthalmology should be carried out within two weeks of injury wherever possible to minimise time to surgery Surgery (if required) should be performed as soon as possible after all required reviews have taken place,9,17–19 ideally within two weeks of injury Other methods of reducing the DNA rate of patients with orbital trauma should be explored to avoid potential adverse effects of delayed treatment
20
and health service resource wastage
35
In conclusion, we believe that our interventions have improved the care of patients at our unit with orbital trauma. We hope that similar strategies could be implemented in other units and within other countries and across a range of specialties with similar effect. We would be interested in obtaining data from other units with outcome data on orbital trauma to assist in the establishment of national guidelines on the management of these patients.
Supplemental Material
Supplemental material for Exploring the effect of streamlining multidisciplinary care pathways on orbital trauma outcomes at a regional referral centre
Supplemental Material for Exploring the effect of streamlining multidisciplinary care pathways on orbital trauma outcomes at a regional referral centre by Matthew Gillam, Niraj Mandal, Vickie Lee, Vihba Vora and Michael Perry in International Journal of Care Coordination
Footnotes
Author’s note
All authors are also affiliated with London Northwest Healthcare NHS Trust, UK.
Acknowledgements
The authors would like to thank Mr. Simon Holmes for his assistance in planning the study.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Supplemental material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
