a b s t r a c t

Background: Ocular trauma is recognized as the leading cause of unilateral blindness. However, few studies to date have focused on the clinical features of hospital-based ocular emergencies. Effectiveness of trauma centers in treating ocular emergencies was compared with treatment in traditional community hospital emergency departments. Demographics, causes, and nature of ocular emergencies, as well as visual outcome in community hospitals emergency departments and trauma centers, were also examined.

Methods: Records of 1027 patients with ocular emergencies seen between July 2007 and November 2010 at 3 community hospitals emergency departments and 2 hospitals with level II trauma centers were retrospectively examined. Unpaired t test and Pearson ?² test were used to determine statistical significance. Results: The incidence of patients requiring ophthalmic intervention was 77.2 per 100 000 in the community hospitals and 208.9 per 100 000 in the trauma centers. Rates of ocular emergencies were higher in middle- aged, white men. Orbital fractures were found in 86% of all orbital contusion cases in trauma centers, whereas 66.7% of patients with fall injuries and open globe diagnoses resulted in legal blindness.

Conclusions: The middle-aged, white men are more vulnerable to ocular injuries caused mainly by motor vehicle accidents. The ability of trauma centers to provide comparable increases in vision outcomes, despite treating more severe ocular emergencies, demonstrates the effectiveness of trauma centers. Patients diagnosed as having orbital contusions or who have fall injuries deserve careful evaluation because they are more likely to have more severe sight-threatening injuries.

(C) 2014

Introduction

Ocular emergencies are a significant component of modern hospital-based emergency health care [1]. The Baltimore Eye Survey reported a cumulative lifetime prevalence of eye injury in 14.4% of the general population [2]. Another study showed an annual rate of hospitalization due to ocular trauma of 13.2 per 100 000 individuals [3]. Furthermore, the incidence of ocular emergencies requiring emergency department (ED) presentation in the United States has been estimated at a national average of approximately 3 per 1000 persons a year, with 2.4 million ocular injuries occurring every year in the United States [4]. Nationally, ocular injuries make up 3.3% of all

? Conflict of interest: The authors declare no conflicts of interest.

?? Authorship: All authors participated in designing this study. C.C., M.M., L.G., and

A.C. contributed in the literature search, Data interpretation, and data analysis. All authors contributed to data collection and the writing of the manuscript. C.C., M.M., L.G.,

A.C., J.B.M., and J.R.M. provided critical revision of the manuscript.

? Disclosure: The authors declare no conflicts of interest.

* Corresponding author. 11216 Trinity River Dr, Rancho Cordova, CA 95670. Tel.: +1 9167126766.

E-mail address: [email protected] (J.R. Martel).

occupational injuries resulting in lost workdays according to National Safety Council, 2002. Ocular trauma is the major cause of unilateral blindness in the United States, affecting 40 000 and 60 000 patients annually [1].

Despite the significance of ocular emergencies in the United States, there has been little literature focused on the nature of ocular emergencies and even less epidemiologic research on the effect of trauma centers (TCs) on Treatment outcomes after traumatic ocular injury. Trauma centers have been developed to presumably improve emergency care services, particularly for those emergencies of traumatic nature that can result in high morbidity and mortality. As such, TCs should represent a concentration of severe ocular emergencies and be able to optimize the outcome of treatment. However, an estimated 50% of all injured patients receive treatment at non-TCs due to reasons such as regional availability, geographic considerations, and practicality of treatment [5]. Therefore, there is a need to understand the potential differences in patient populations as well as treatments and outcomes between community hospitals (CHs) with non-TC EDs and those that have certified TCs.

We performed a retrospective study to evaluate the epidemiology and clinical features of hospital-based serious ocular emergencies,

0735-6757/$ - see front matter (C) 2014 http://dx.doi.org/10.1016/j.ajem.2013.11.015

222 C.A. Cheung et al. / American Journal of Emergency Medicine 32 (2014) 221-224

Table 1

Patient demographics within TCs and CHs based on age, sex, and race

Characteristics No. of patients (n = 1027) TCs (n = 785) CHs (n = 242) Age (y), n (%)

b18	189 (11)	77 (9.8)	35 (14)
18-30	214 (21)	176 (22)	38 (16)
31-44	170 (17)	135 (17)	35 (14)
45-64	310 (30)	240 (31)	70 (29)
>=65	220 (21)	156 (20)	64 (26)
Sex, n (%) Male	612 (60)	485 (62)	127 (52)
Female	415 (40)	300 (38)	115 (48)
Race, n (%)? White	710 (69)	565 (72)	145 (60)
Hispanic	108 (11)	73 (9.2)	35 (14)
African American	87 (8.5)	57 (7.3)	30 (12)

Table 3

Cause of injury rates within TCs and CHs

By percentage (n = 1027)

Cause? MVAs Fall Infection Preexisting

				condition			unknown source
Overall population	17.4	14.4	14.3	14	11.5	11.1	7.5
TCs	22.8	15.4	9.9	11.3	10.6	13.2	6.9
CHs	0	11.1	28.5	22.7	14.5	4.1	9.5
Sex Male	19	13.2	11	11.6	8.5	15.9	8.6
Female	15.1	16.14	19.3	17.6	15.9	4.1	5.8

Other⁺ Assault Trauma

* We were able to obtain information regarding race for 966 patients. There were a number of patients who either refused to provide racial information for personal reasons or were too ill to do so.

critical enough to warrant an ophthalmology consult, from 5 Regional hospitals over a 3-year period. Three of these hospitals were traditional CH EDs, and 2 hospitals had well-established level II TCs. In this study, we determined the effectiveness of TCs in treating ocular emergencies as well as delineated differences between the de- mographics of patient populations, causes and nature of ocular emergencies, treatment, and outcome in CHs and TCs.

Methods

Medical records of hospital-based ocular emergencies for 5 hospitals during the period beginning in July 2007 and ending in November 2010 were retrospectively examined. Only hospital-based ocular emergencies that were severe enough to warrant an ophthal- mology consult were included. Less severe ocular emergencies that did not necessitate initial direct ophthalmology involvement or required only ophthalmology follow-up were not included in our study.

We reviewed patient demographics, location of emergency, cause of ocular emergency, diagnoses, treatment, and treatment outcome. Patients’ ages were grouped according to the following categories:

0 to 17 years (children and adolescents), 18 to 30 years (early

adulthood), 31 to 44 years (early middle age), 45 to 64 years (middle age), and 65 years and older (elderly).

Owing to its retrospective nature, the study was exempt from informed consent by our local institutional review board committee. We used SPSS (version 19; SPSS, Chicago, IL) for statistical analysis of data. Unpaired t test and Pearson ?² test were used to determine statistical significance, which was considered to be a 2-tailed P value less than .05.

Results

During the period from July 2007 to November 2010, the 3 CHs treated a total of 312 215 emergency patients and the 2 TCs treated 375 711 patients. There were 1062 patients with hospital-based ocular emergencies. We excluded 35 patients because of insufficient data in their medical records, resulting in a final sample size of 1027

	Diagnosis category	Total diagnosis	TC	CH	Male	Female
	Contusion	25.0	27.6	13.9	27	21.6
	Orbital trauma	18.1	20.6	7.3	19.7	15.3
	Posterior segment	15.1	15	16.3	15.2	15.2
	Anterior segment	9.7	9	13	10.1	9.2
	Neurologic	9.7	8.1	13.6	7.9	12.3
	Infections	8	5.4	19.8	5.9	11.8
	Adnexal	7.8	8	7.6	7.8	8
	Open globe	3.3	2.9	5.4	5.9	2.6
	Foreign body	2	2.1	1.4	7.8	2.3
	Glaucoma	1.3	1.3	1.6	3.7	1.6

Table 2

age distribution based on sex By percentage (n = 1027)

	Age (y)	Male	Female	All patients
	b18	11.8	9.6	10.9
	18-30	21.9	19.3	20.9
	31-44	19.1	12.8	16.6
	45-64	31.3	28.7	30.2
	>=65	15.9	29.6	21.4

* Seven of the most common causes of ocular emergencies in this study.

Age (y) b18	10.7	10.7	22.3	6.3	7.1	0	12.5
18-30	28	7.5	10.7	7.5	7.9	21	8.4
31-44	15.9	10.6	13.5	11.8	9.4	15.9	6.5
45-64	20	13.2	13.2	16.5	9.7	11.9	6.5
>=65	8.2	27.7	15.9	22.7	15.9	1.8	5.9

⁺ Including burns, explosion, gunshot, pellet guns.

patients with 2025 ocular emergency diagnoses. Trauma centers accounted for 76.4% of hospital-based ocular emergencies, whereas CHs reported 23.6% of the ocular emergencies. On average, there were

2.1 ocular diagnoses per patient in TCs compared with 1.5 diagnoses in CHs. The incidence of patients requiring ophthalmic intervention was 208.9 per 100 000 in TCs and 77.2 per 100 000 in CHs. Many of the hospital-based ocular emergency patients had multiple injuries or comorbidities.

Demographics

Demographic data were analyzed according to age, sex, and race (Table 1). In our study population, ocular emergencies were more prevalent in middle-aged, white men. The rates of eye injuries were found to be greater in patients in the age range of 45 to 64 years, followed by those 65 years or older and those within the age range of 18 to 30 years. The mean age in TCs was 41.7 years for men and 49.7 years for women (P <= .001; Table 2). The mean age in CHs was 47.2 years for men and 47.1 years for women (P = .99). Furthermore, the TC patient population was significantly more male predominant compared with that of CHs, suggesting that men are involved with more severe eye injuries, requiring specialized trauma intervention. injury rates were highest among whites in both the TCs and CHs populations.

Ocular emergencies

Ocular emergencies were analyzed according to the cause, location of injury, diagnoses, treatment type, and Visual outcomes based on visual acuity. The most common cause of hospital-based ocular emergencies was motor vehicles accidents (MVAs), which accounted

Table 4

Diagnoses of ocular emergencies within TCs and CHs

By percentage (n = 1027)

C.A. Cheung et al. / American Journal of Emergency Medicine 32 (2014) 221-224 223

Table 5

Visual acuity change by hospital type

	By percentage (n = 301)
	Vision	All patients	CH	TC
	Increased	62.1	57.5	63.8
	Same	17.3	20	16.3
	Decreased	15.3	18.8	14
	Total blindness (NLP)	5.3	3.8	5.9

NLP, No light perception.

for 17.4% of all ocular emergencies, followed by fall-related accidents and infectious agents (Table 3). Male emergencies were most commonly caused by MVAs, whereas most female ocular emergencies were a result of infectious agents. Traumatic ocular emergencies accounted for 58.3% of all ocular emergencies in TCs compared with 24.7% in CHs. Comparisons between hospital-based ocular emergen- cies yielded statistically significant differences in regard to MVAs, infectious agents, preexisting conditions, and assaults (P b .05).

Of patients’ location of injury obtained (n = 480), most injuries occurred on the street and highways (47.7%), followed by injuries that occurred at home (36.7%). The differences in location of injury based on hospital type also reached statistical significance (P b .05) in regard to injuries that occurred on the streets and highways (TCs 55.2%; CHs 9%) and those injuries that occurred at home (TCs 28.9%; CHs 76.3%). The most common diagnoses of hospital-based ocular emergencies were ocular contusions, orbital trauma, and posterior segment disease (Table 4). Both TC and CH differences between contusion, anterior segment disease, open globe injuries, orbital trauma, infectious disease, and neurologic diagnoses were all statistically significant (P b .05). Men had statistically significantly more contusions and orbital trauma (P b .05). Women had statistically significantly more

infections and neurologic emergencies (P b .05).

Of all ocular emergencies, 54.0% were medically treated, 39.4% required surgical intervention, and 6.5% required only observation or further outpatient testing. The differences in treatment of ocular emergencies based on hospital type also reached statistical signifi- cance (P b .05) in regard to surgical interventions (TCs 41.7%; CHs 32.2%) and observational treatment (TCs 5.2%; CHs 10.7%). The number of cases that required medical treatment between the hospital types was not statistically significant (TCs 53.1%; CHs 57.0%). Outcomes were also evaluated by initial and final visual acuity (Table 5). Owing to the transitory nature of ED visits, outcome data for 301 patients were acquired, with 221 patient outcomes from TCs and 80 from CHs. Of patients whose outcome data were obtained, TCs had a greater rate of improved visual acuity after treatment but also an increased rate of blindness in comparison with CHs. Of the CH patient outcome data obtained, 57.5% had increased vision and 3.8% resulted in blindness. However, none of the outcome differences between the hospital types reached statistical significance (P = .05). Final visual acuity increased in 60.5% of the surgically treated patients and

Table 6

Rates of blindness from open globe and orbital fracture injuries, sustained from MVAs and fall injuries

decreased in 17.9%. For the medically treated patients, 65.4% experienced an increase in visual acuity, whereas 12.6% experienced a decrease (P N .05).

Visual outcomes for the 2 most common causes of ocular trauma, MVAs, and fall injuries were examined (Table 6). Specifically, we examined injuries that resulted in legal blindness.

Discussion

In the United States, the incidence of ocular emergencies has varied in previous studies, from 27 to 490 injuries per 100 000 [4]. We observed a significantly higher incidence of ocular emergencies in TCs, with 208.9 per 100 000 cases compared with 77.2 per 100 000 cases in CHs. The 2 TCs accounted for 76.4% compared with 3 CHs, which accounted for 23.6% of ocular emergencies. The complexity of ocular emergencies also appeared to be greater in TCs, with a higher average number of diagnoses per patient. Trauma centers had a greater number of trauma-related ocular emergencies, whereas CHs had more infection-related cases. In addition, TCs had a statistically greater incidence of cases requiring surgical treatment. This would suggest that our sample region is using the trauma care system effectively and appropriately distributing patient emergencies by severity and cause of injury.

In both TCs and CHs, ocular emergencies were most common in middle-aged (45-64 years), white men, which is in agreement with other studies [1,4,6,7]. The age results of our study were also in agreement with many previous studies, which indicated that eye injuries are prevalent in younger population [6,8,9].

Among current literature, an increased incidence of hospitalized ocular emergencies has also been noted for the elderly [8]. In our study, we noted that 21% of all ocular emergencies were found in the elderly, with most of the injuries being caused by falls. In our study, fall injuries were most common in elderly women, but other reports describe a plateau in the sex-based difference at this age range [3,9,10], whereas others found continued higher prevalence of ocular emergencies among older men [11,12]. The changing male-to-female ratios within the elderly population could be explained by the decrease in men in the elderly population or possibly by the increasingly similar lifestyles between men and women as they age [12].

We found that the most common cause of ocular emergencies overall was traumatic in nature, predominantly as a result of MVAs and fall injuries. A 4-year study of the National Trauma Databank correlated with our results, citing MVAs as the most frequent mechanism of injury among ocular injury patients per year [1]. Because MVAs frequently result in multiple injuries, knowledge of the common concomitant injuries may guide ED staff to more timely diagnoses and treatment, limiting ocular morbidities, which affect quality of life.

Orbital contusion was the etiology of 27.6% of the cases seen in TCs compared with 13.9% in CHs. Remarkably, orbital fractures were found in 86% of all orbital contusion cases in TCs. Therefore, emergency personnel in a TC environment should rule out orbital fractures in patients with suspicious presentation of orbital contusion. We found no statistical significance between the visual outcomes between cases treated in CHs and the more severe cases treated in

Variable Total no. of patients % of diagnosis in

patient population

% of patients with final visual acuity of 20/200 or worse (legal blindness)

TCs. The ability of TCs to provide comparable increases in vision

outcomes, despite treating more severe ocular emergencies, demon- strates the effectiveness of TCs. Although we anticipated that the most severe ocular injuries would occur in MVAs, often with relative speeds greater than 60 miles/h, we found that fall injuries among the elderly resulted in poorer prognosis. Comparably, 45.3% of fall injuries and 46.9% of MVAs resulted in orbital fractures. However, only 1.1% of the MVA emergencies resulted in open globe injuries as compared with the 6.1% of fall injuries. More importantly, 66.7% of the fall patients with open globe injuries resulted in legal blindness as compared with the MVA injuries, which in resulted in no cases of legal blindness. We

Fall injury Open globe	148	6.1	66.7
Orbital fracture		45.3	1.5
Both MVA	179	1.4	50
Open globe		1.1	0
Orbital fracture		46.9	0
Both		0	0

224 C.A. Cheung et al. / American Journal of Emergency Medicine 32 (2014) 221-224

suspect that air bag deployment in MVA injuries, although associated with ocular injuries, also appears to provide protection from the more severe ocular trauma, particularly when combined with a seat belt restraint. Although we did not further investigate the details of MVA mechanisms of ocular injury, a 2008 study had shown that Airbag deployment alone has been associated with an increased incidence of ocular injuries. However, when used in conjunction with seatbelts, there is a decrease in severe ocular injuries and improved visual acuity outcomes [13]. Therefore, ED personnel in TCs and community centers should be aware of the high risk of orbital fractures and open globe injuries sustained from fall injuries, which appears to be the most common cause of legal blindness.

This study has several potential limitations owing to the study’s retrospective nature, such as lack of some information in patients’ records, as some patients were intubated or unconcious and not able to provide demographics or subjective examination information. Also, owing to the small sample of visual outcomes from CHs obtained, there may have been a limitation in power to detect statistical significance between the visual outcomes in TCs and CHs.

In summary, we presented a comprehensive study of ED-related eye emergencies within this region, including demographics location and nature of the ocular emergency, treatment, and outcome based on hospital type. Results of the study indicate that the middle-aged, white men are more vulnerable to ocular injuries, caused mainly by MVAs. Trauma centers have proved to be effective in treating ocular injuries and improving visual outcomes. Results of the study indicate the subpopulations of patients who present with orbital contusion or fall injury deserve particularly careful evaluation because they are more likely to have more severe sight-threatening ocular injuries.

References

1. Scruggs D, Scruggs R, Stukenborg G, Netland PA, Calland JF. Ocular injuries in trauma patients: an analysis of 28,340 trauma admissions in the 2003-2007 National Trauma Data Bank National Sample Program. J Trauma Acute Care Surg 2012;73(5):1308-12.
2. Katz J, Tielsch JM. Lifetime prevalence of ocular injuries from the Baltimore eye survey. Arch Ophthalmol 1993;111:1564-8.
3. Klopfer J, Tielsch JM, Vitale S, et al. Ocular trauma in the United States: eye injuries

   resulting in hospitalization, 1984 through 1987. Arch Ophthalmol 1992;110:

   838-42.

   McGwin Jr G, Owsley C. Incidence of emergency department-treated eye injury in the United States. Arch Ophthalmol 2005;123:662-6.

4. Mohan D, Barnato AE, Rosengart MR, et al. trauma triage in the emergency departments of nontrauma centers: an analysis of individual physician caseload on triage patterns. J Trauma Acute Care Surg 2013;74(6):1541-7.
5. May DR, Kuhn FP, Morris RE, et al. The epidemiology of serious eye injuries from the United States eye injury registry. Graefe’s Arch Clin Exp Ophthalmol 2000;238: 153-7.
6. 2009 Eye Injury Snapshot Project Results. American Academy of Ophthalmology. Available at: http://www.aao.org/practice_mgmt/eyesmart/snapshot_2009_ results.cfm. Accessed February 4, 2011.
7. Tielsch JM, Parver L, Shankar B. Time trends in the incidence of hospitalized ocular trauma. Arch Ophthalmol 1989;107:519-23.
8. Wilson MR, Wooten F, Williams J. frequency and characteristics of ocular trauma in an urban population. J Natl Med Assoc 1991;83:697-702.
9. Morris RE, Witherspoon CD, Helm HA, et al. Eye injury registry of Alabama (preliminary report): demographics and prognosis of severe eye injury. South Med J 1987;80:810-6.
10. People quick facts. US Census Bureau State and County QuickFacts. Available at: http://quickfacts.census.gov/qfd/states/06/06067.html. Accessed February 4, 2011.
11. Kim JG, Linakis JG, Mello MJ, Armstrong GW, Greenberg PB. Eye injuries in elderly adults presenting to U.S. emergency departments. J Am Geriatr Soc 2013;61(3): 470-2.
12. Rao SK, Greenberg PB, Filippopoulos T, Scott IU, Katsoulakis NP, Enzer YR. Potential impact of seatbelt use on the spectrum of ocular injuries and visual acuity outcomes after motor vehicle accidents with airbag deployment. Ophthalmology 2008;115(3):573-6.e1.