Quantifying Eye Alignment in Orbital Fracture Patients: The Digital Hess Screen

Abstract

Background: Quantifying diplopia to determine management and track outcomes for orbital fracture patients is vital for standardization between visits, physicians, and coordination among the multiple specialties that manage these patients. However, standardization is challenging, as diplopia is often reported subjectively. This study sought to describe the utility of the digital Hess screen in patients with orbital fractures compared with a control group.

Materials and Methods: A prospective pilot study was designed in which adult patients who presented with orbital fractures between November 2017 and January 2019 without prior history of orbital pathology were recruited. Subjects underwent digital Hess screen testing, in which they wore anaglyph glasses and aligned targets on a computer screen to quantify static eye alignment. The degree of any eye misalignment was analyzed and compared with controls.

Results: Ninety-one patients and 35 controls were enrolled. All participants were able to complete the digital Hess screen. Average cumulative deviation score of orbital fracture patients within 1 month of injury was 0.65°, compared with 0.28° in controls. This was a statistically significant difference (p < 0.01, 95% confidence interval −0.18 to 0.18).

Conclusion: The Hess screen has been used to quantify phoria as a correlate of eye alignment and diplopia, but older versions were cumbersome and difficult to analyze. This study is the first to report on using the digital Hess screen to quantify phoria in orbital fracture patients and provides a more concise and standardized means to track clinical and surgical outcomes of eye alignment.

Key Points

Question: Can the digital Hess screen be used to quantify eye alignment in orbital fracture patients?

Findings: All orbital fracture patients recruited were able to complete the digital Hess screen test, and had significantly higher eye misalignment compared with control subjects.

Meaning: This study is the first to report on using the digital Hess screen to quantify phoria in orbital fracture patients and provides a more concise and standardized means to track clinical and surgical outcomes of eye alignment.

Introduction

Orbital fractures are one of the most commonly encountered facial fractures.^1,2 Functional and structural deficits are common with orbital fractures, and can result in subsequent diplopia and enophthalmos due to altered position of the globe and extraocular muscle pulley system.² Major visual impairment and undesirable aesthetic sequelae can persist if not treated appropriately.^3–5 Despite their common nature and significant associated morbidity, the outcome measures after clinical observation and surgical management of orbital fractures remain highly varied.

Currently, assessment of orbital fracture patients includes radiographic imaging and clinical examination for signs of extraocular muscle entrapment, fracture size, and evidence of diplopia and enophthalmos.^6,7 Patients are classified based on aesthetic and functional concerns, including patient-reported diplopia, enophthalmos, proptosis, restricted extraocular movements, strabismus, and infraorbital hypesthesia.^8–10

Quantifiable parameters to guide reconstruction are becoming more popular, and recent studies in reconstruction outcomes are utilizing 3D computed tomography (CT) to quantify changes in orbital volume¹¹ and globe protrusion pre- and postsurgery.¹² Quantifiable parameters of clinical outcomes have been more difficult to standardize for orbital fractures. Establishing objective metrics in the evaluation and treatment of ocular misalignment encountered with orbital fractures is important in improving patient outcomes.

The Hess screen is a test of ocular motility.¹³ The subject wears anaglyph glasses with a red and a blue lens, and sees a red and blue target on the screen. The wavelength of each respective target is canceled out by the tint of the glasses such that the right eye sees only the red target and the left eye sees only the blue target. In this way, binocular vergence input is eliminated and eye alignment, specifically phoria, can be measured and quantified. Studies examining ocular alignment after orbital fracture repair have used the original Hess screen to quantify outcomes with varying results.^10,14–16

Although the original test offered a way to quantify eye alignment for orbital fracture patients, the Hess screen has not previously gained widespread use likely because the original test required cumbersome equipment, including a physical pointer and large chart as well as a technically skilled test administrator.¹³ However, in recent years the Hess screen has evolved into a digital computer program in which the patient uses the computer mouse to align a blue target with a red target and vice versa for each eye across 50 targets with a maximum of 16° around central gaze for each eye. The degree of target alignment is then quantified within the program, giving a specific degree of deviation output that can be averaged from the horizontal and vertical deviation of each target point. The test also typically takes <5 min to complete if a subject has previous experience using a computer and computer mouse.

The objective in this study was to employ the digitized Hess screen to quantify the effect of orbital fracture on eye alignment. We hypothesized that the digital Hess screen could be easily executed in all orbital fracture patients, and that patients with Hess screen results that demonstrated greater eye misalignment would correlate with patients who reported diplopia.

Materials and Methods

A prospective pilot study was approved by the University of Washington Institutional Review Board. Subjects recruited included all adult patients who presented to our institution's otolaryngology and craniofacial clinics for assessment of a traumatic orbital fracture. Concurrently, control subjects were recruited by approaching any available adult >18 years who did not have any prior history of ocular motility disorder or history of ocular pathology. The study period was from November 2017 to January 2019.

Subjects were recruited at their initial outpatient visit to the otolaryngology or craniofacial plastics clinics and met inclusion criteria if they were >18 years, had a diagnosed orbital fracture confirmed by CT scan within 1 month of recruitment, and were physically able to open both eyes such that they could view a computer screen. Subjects were excluded if they had any history of prior orbital fracture, history of prior ocular motility disorder, including pre-existing diplopia, history of previous ocular motility surgery, or history of orbital reconstruction. After informed consent, each subject underwent digital Hess screen testing with the PC Hess Screen (Thompson Software Solutions, City University, London, UK) at the time of enrollment within 1 month of injury and before any anticipated reconstruction.

Testing was performed in the following manner: the subject sat in front of the computer screen wearing anaglyph (red–green lens) glasses (Fig. 1). The distance from the subject to the screen was consistently measured at 40 cm to maintain consistency. The patient was seated such that his or her eyes were level with the center point of the testing screen. A red and blue target were then displayed, and the subject had to use the mouse to attempt to align the red and blue target, clicking the mouse control once perceived alignment was achieved. This task did require that the subject was able to understand and perform the computer alignment task that restricted testing to individuals able to understand the task goals in English and those with an ability to use a computer mouse to make their target selections. Once the patient selected a target, the computer registered the relative position of the target and stimulus. This task was repeated for 25 grid points for fixation of the left eye, and then repeated for another 25 grid points for fixation of the right eye by reversing the color of the fixed point and movable target.

Fig. 1.

Digital Hess screen testing. (A) A control subject undergoing Hess screen testing. (B) The goal of the tester, to place the single blue line inside the two red lines. (C) The left eye only sees the blue line, whereas (D) shows that the right eye only sees the red lines.

Test results were plotted on a digital Hess screen chart as well as tabulated in an exportable .txt file. This file was imported into Excel. The Hess area ratio (HAR) was calculated as a comparison with reports of the classic Hess screen output in prior studies. This was calculated by measuring the length of inter-horizontal plot and inter-vertical plot on the central line of the Hess chart and calculating the percentage of square area of the affected side compared with the healthy side.¹⁷ HAR follows the equation HAR (%) = 100 × (A × B)/(A′ × B′) (%), where A is the affected side length between horizontal plots, B is the affected side length between vertical plots, A′ is the healthy side length between horizontal plots, and B′ is the healthy side length between vertical plots. For control subjects, the side with the greatest deviation was considered the affected side. The average degree of deviation was then calculated by summating the absolute degree difference between the subject-selected target and the true target in vertical and horizontal directions, averaged across 50 target points in a grid ranging 16° above, below, left, and right from central gaze (Fig. 2).

Fig. 2.

Digital Hess screen plot of orbital fracture subject. Target selections when each is fixated on the static target and the contralateral eye is tracking the marker using the computer mouse.

Data collection involved recording each subject's demographic information, date of orbital fracture injury, and Hess screen data. Exact time since injury was recorded for each patient at the time of enrollment. Given the pilot nature of this study, no sample size was calculated to guide enrollment goals as no previous reports on digital Hess screen measurements in this patient group exist to our knowledge. Standard descriptive statistics were calculated to identify ranges, distributions, means, and standard deviations of each demographic point and the digital Hess screen scores. Statistical analysis was performed in consultation with a biostatistician through the University of Washington, Department of Ophthalmology. The primary outcome was the change in digital Hess screen average point deviation between controls and orbital fracture patients. The secondary outcome was the change in HAR.¹⁷ A p-value of ≤0.05 was considered significant for all analyses.

Results

Ninety-one orbital fracture patients and 35 controls were enrolled in the study. Demographics of the patient population are described in Table 1. The majority of patients were male (70%) and had an average age of 42 years. The mechanism of injury was approximately even between being struck (37%), having a fall (34%), or having a vehicle-related injury such as a motor vehicle or bicycle collision (29%). Eighty-five percent of patients had an orbital floor fracture and 45% had a lateral wall fracture, followed by 36% with a medial wall and 12% with a superior wall fracture. Fifty-two percent of patients had two-wall fractures.

Table 1.

Demographics of orbital fracture subjects

Variable	n (% total)
Age	42 years
Gender
Male	64 (70)
Female	27 (30)
Average time from injury to testing	10 days
Mechanism
Struck	34 (37)
Fall	31 (34)
Vehicle related	26 (29)
Fracture type
Lateral	41 (45)
Medial	33 (36)
Floor	77 (85)
Superior	11 (12)
Number of orbital walls fractured
1	35 (38)
2	47 (52)
3	5 (5)
4	3 (3)
5	1 (1)
Involving other facial bones
ZMC	32 (35)
NOE	6 (7)
LeFort	4 (4)
Frontal sinus	1 (1)
Subjective diplopia
Yes	16 (18)
No	75 (82)

NOE, naso-orbital ethmoid; ZMC, zygomaticomaxillary complex.

Additional facial fractures were present in 47% of patients, with the majority of these having a zygomaticomaxillary complex fracture. Although 18% of all fracture patients reported diplopia on their initial otolaryngological clinical examination, only 43% of these patients had a formal ophthalmology evaluation. Thus, only 38% of patients reporting subjective diplopia were confirmed by an ophthalmologist to have diplopia specifically related to ocular motility.

The HAR was calculated for each control and patient (Fig. 3). There was no significant difference between the HAR for the control subjects and fracture subjects: the average HAR for control subjects was 0.99, and the average HAR for fracture patients was 1.00.

Fig. 3.

Hess area ratio: controls versus all fracture patients. Box plot displaying the quartiles, means (X), and outlier subjects as points above and below each quartile.

The average point deviation median of the control subjects was 0.22°, compared with the average point deviation median of the orbital fracture patients that was 0.39° (Fig. 4). Mann–Whitney U test gave a statistically significant p-value of <0.01 demonstrating that the medians between the control and subject average point deviations was statically significant. The average point deviation mean for the 35 control subjects was 0.28°, whereas the average point deviation mean for the orbital fracture patients was 0.65 (Fig. 4). In a two sample t-test comparison of means, the p-value was also statistically significant at <0.01 (95% confidence interval [CI] −0.18 to 0.18).

Fig. 4.

Hess screen average point deviation: controls versus all fracture patients. Box plot displaying the quartiles, means (X), and outlier subjects as points above and below each quartile.

When orbital fracture patients were further subdivided into patients who did not report subjective diplopia and those who did, the average point deviation median for the 75 patients who did not have subjective diplopia was 0.33°, compared with 0.93° for the 16 patients who did report subjective diplopia (Fig. 5). Mann–Whitney U test comparison of medians between the controls and the patients without diplopia gave a p-value of <0.01 between the average point deviation median. Similarly, the average point deviation mean for the 75 patients who did not have subjective diplopia was 0.51°, compared with 1.30° for the 16 patients who did report subjective diplopia. This was also a statistically significant difference with p = 0.02 in a two sample t-test comparison of means (95% CI difference of means −0.68 to 0.68).

Fig. 5.

Hess screen average point deviation: controls versus patients with and without subjective diplopia. Box plot displaying the quartiles, means (X), and outlier subjects as points above and below each quartile.

Discussion

Quantitative functional measurements of orbital fracture patients are necessary to standardize approach and management. Historically, functional ocular motor testing has been reported using the Hess screen. However, in its original state, the Hess screen is cumbersome to perform and requires technical expertise. As a result, computerized forms of the test have been developed in the past several decades.

The digital or computer Hess screen has previously been described in the literature in direct comparison with the conventional Hess screen for patients with general diplopia.¹⁸ Although initial studies found that right eye horizontal deviations were larger with the digital Hess screen, and left eye vertical deviations were smaller, the ultimate conclusions were that due to the ergonomic advantages and overall equivalent performance the digital Hess screen remained superior and replaced conventional testing in the researchers' practices. The digital Hess screen has furthermore started to gain traction as an outcome measure in surgical studies, specifically measuring eye alignment before and after deep lateral wall orbital decompression in thyroid-related orbitopathy.¹⁹ Researchers showed an evolved reporting approach by using the amplitudes of deviation, made significantly easier as a result of the computer-generated report after testing.

This study applies the digital Hess screen as a computerized quantifiable measurement system of ocular motility and alignment after initial orbital trauma. A total of 91 fracture patients and 35 controls were enrolled, and all subjects were able to complete the digital Hess screen after simple instructions were given. This demonstrates the ease of use of this test in a trauma patient population, along with broad applicability to subjects ranging in age from 18 to 83 years. This study is also unique in that we recruited both orbital fracture patients and a control group to have a specific baseline from which to compare fracture patient results.

Data in this study are reported in both HAR similar to classic studies, as well as in average point deviation to account for all 50 plotted point horizontal and vertical misalignments. Although the average HAR was not significantly different between controls and fracture patients as a whole (0.99 compared with 1.00, respectively), the average point deviation showed a large statistically significant difference between these groups in both mean and median (0.28 compared with 0.65 and 0.22 compared with 0.39). The average point deviation was, therefore, able to detect ocular misalignment differences on a more detailed level once each grid point horizontal and vertical deviation was taken into account.

There was a significant difference in measured eye alignment between groups of patients who reported diplopia and those who did not. Within the orbital fracture patient cohort, although the average point deviation was as a whole higher than those of controls, when patients were separated by diplopia status there remained those who had significant eye misalignment in both groups: 9 of 75 patients (12%) in the no subjective diplopia group and 7 of 16 patients (44%) in the subjective diplopia group had more than one degree of deviation on average. This implies that even in patients who do not report diplopia after orbital fracture, eye misalignment can be quantified and significantly differ compared with control subjects. Conversely, a subset of patients who reported subjective diplopia was found to have normal eye alignment on digital Hess screen testing. Findings such as this highlight the need to classify subjective diplopia through quantitative measures, as subjective diplopia may not correlate with quantifiable eye alignment discrepancies but instead may relate to other ocular pathologies, including damage to the retina, cornea, or lens that would not be improved through orbital fracture repair alone.

Limitations to this study primarily include the heterogenous nature of the orbital fracture patient population, with a wide variety of fractures represented. The diversity of fractures and traumatic mechanisms of injury may have a significant effect on eye alignment, potentially masking more specific patterns of eye alignment deviation in certain types of fractures or injury mechanisms. Although patients were tested within a month of injury, the range of time tested since injury was between 3 and 28 days and could represent an additional source of heterogeneity. Furthermore, the majority of patients were not concurrently evaluated by an ophthalmologist to determine whether their reported subjective diplopia may be due to independent ocular pathology. Future studies may focus on direct comparisons of diplopia reported by patients, ophthalmologist assessment, and digital Hess screen findings to create an overall quantifiable classification scheme when reporting orbital fracture outcomes data.

Conclusion

Quantifying diplopia to determine management and track outcomes for orbital fractures is challenging, as it is often reported subjectively. The Hess screen has been used to quantify phoria as a correlate of eye alignment and diplopia, but older versions were cumbersome and difficult to analyze. This study is the first to report on using the digital Hess screen to quantify phoria in orbital fracture patients. It demonstrates the feasibility and ease of use of this testing modality for this patient population. In addition, these data show that a subset of patients who did not report subjective diplopia have measurable abnormalities in their eye alignment, which may affect their clinical course. As orbital fracture management is undertaken by a wide group of disciplines, including otolaryngology, ophthalmology, plastic surgery, and oral maxillofacial surgery, the use of the digital Hess screen to quantify eye alignment outcomes has the potential to cross many specialties and establish a greater standard management approach.

Ethical Approval

This study was approved by the University of Washington Institutional Review Board (study no. 00000999).

Author Disclosure Statement

No competing financial interests exist.

Funding Information

The time provided by S.R.A. was supported by the United States National Institute on Deafness and Other Communication Disorders under grant T32 DC000018. Additional support was provided through the United States National Institute of Health CORE grant EY001730, an unrestricted grant from Research to Prevent Blindness.