Authors: Kenny Y. Wang (1.Alix School of Medicine, Mayo Clinic, Rochester, Minnesota), B. Michelle Kim (1.Alix School of Medicine, Mayo Clinic, Rochester, Minnesota), Timothy T. Xu (2.Department of Ophthalmology, Mayo Clinic, Rochester, Minnesota), Margaret M. Reynolds (2.Department of Ophthalmology, Mayo Clinic, Rochester, Minnesota; 3.Department of Ophthalmology and Visual Sciences, Washington University School of Medicine, St. Louis, Missouri), David O. Hodge (4.Department of Quantitative Health Sciences, Mayo Clinic, Jacksonville, FL), Wendy M. Smith (2.Department of Ophthalmology, Mayo Clinic, Rochester, Minnesota)
Categories: Article, Traumatic iritis, Uveitis, Epidemiology, Incidence, Mechanisms
Source: Ocular immunology and inflammation
Authors: Kenny Y. Wang, B. Michelle Kim, Timothy T. Xu, Margaret M. Reynolds, David O. Hodge, Wendy M. Smith
To describe the incidence, mechanisms, and clinical characteristics of patients diagnosed with traumatic iritis in a U.S. Midwestern county population.
Retrospective population-based cohort of all residents of Olmsted County, Minnesota diagnosed with traumatic iritis from January 1, 2006, to December 31, 2015. The medical records of patients with traumatic iritis were identified using the Rochester Epidemiology Project database, which contains virtually all medical care in the county. Medical records were reviewed for demographics, presentation, and follow-up data. Incidence rates were calculated per 100,000 per year.
There were 156 incident diagnoses of traumatic iritis during the 10-year study period, yielding an age- and sex-adjusted incidence rate of 10.7 per 100,000 per year. Traumatic iritis disproportionately occurred in male (p<0.001) and Black (p<0.001) patients. The mean age of diagnosis was 33 years (range: 4–96 years), mean number of traumatic iritis-specific follow-up visits was 2.1 (range: 0–26), and median duration of traumatic iritis-specific follow-up was 11 days (range: 1 day-1.6 years). There were 155 (99.4%) patients with unilateral disease. The most frequent mechanisms of traumatic iritis were sports-related (N=29, 18.6%), assault-related (N=23, 14.7%), scratch (N=22, 14.1%), and work-related (N=21, 13.5%) injuries. The mean initial and final best-corrected visual acuity (BCVA) of the affected eye was 20/40 and 20/30, respectively. Loss of follow-up was more frequently observed in Black patients (p<0.001) and patients with smoking history (*p=*0.004).
Traumatic iritis was most frequently observed in younger males and Black patients. Common mechanisms included sports, assault, scratch, and work-related injuries.
Traumatic iritis is a subtype of anterior uveitis defined as uveal inflammation typically caused by blunt injury. Most common previously reported mechanisms of traumatic iritis include motor vehicle accidents, air soft gun projectiles, and work accidents.^1–3^ In children, traumatic iritis accounts for up to 25% of pediatric uveitis diagnoses.^4,5^
Few studies to date have described the epidemiology and clinical outcomes of traumatic iritis. Most studies of traumatic iritis are referral center-based, potentially introducing selection bias and overrepresenting more severe cases of traumatic iritis.^3,6^ This study aimed to investigate the population-based incidence, mechanisms, and clinical characteristics of traumatic iritis in a single Midwestern U.S. population over a 10-year period using the Rochester Epidemiology Project (REP) database, a database which captures the medical records of nearly all residents of Olmsted County, Minnesota.
This was a retrospective cohort study of all patients in Olmsted County, Minnesota diagnosed with traumatic iritis within a 10-year period from January 1, 2006, to December 31, 2015. Patients were identified using the REP, a medical records database of over 150,000 unique residents of Olmsted County, Minnesota, that tracks medical care administered by Mayo Clinic, Olmsted Medical Group, and small independent clinics.^7,8^ The REP captures virtually all medical care delivered to residents of Olmsted County.^9^ This retrospective cohort study was approved by the Mayo Clinic and Olmsted Medical Center Institutional Review Boards. All REP participants provided authorization for minimal risk research prior to entrance into the database. This study adhered to the tenets of the Declaration of Helsinki and the Health Insurance Portability and Accountability Act. Since this was considered minimal risk, no additional consent from patients was required by the Institutional Review Board of Mayo Clinic and waiver of consent was granted.
Patients residing in Olmsted County, Minnesota from January 1, 2006, to December 31, 2015, who were diagnosed with traumatic iritis were identified with an extensive REP search of International Classification of Diseases, Ninth and Tenth Revisions (ICD-9 and ICD-10) codes for ocular inflammation. All patients with an incident diagnosis of traumatic iritis were included in this study. Exclusion criteria included diagnosis outside the study period and unknown diagnosis date. Research authorizations were confirmed through the database, and patients who did not give authorization were excluded. This investigation was part of a larger study on uveitis epidemiology in Olmsted County, Minnesota, which has previously been reported.^10^ Medical records for each patient were reviewed for demographics, presenting symptoms, treatment, and follow-up. Incident cases were defined as the first encounter in which patients had documented ophthalmic findings consistent with uveitis defined by the Standardization of Uveitis Nomenclature (SUN) criteria.^11^ Cell and flare grading was based on the SUN grading scheme, where “trace” was indicative of 0.5+.^11^ When cell or flare grades were given as a range, the midpoint was used (e.g., 1.5+ for 1–2+). Exam findings with qualitative findings such as “mild” or “significant” were not quantified and therefore not reported. Iritis-specific follow-up was defined as a follow-up visit for traumatic iritis or for a sequela of iritis, whereas general follow-up was defined as any optometry or ophthalmology follow-up in which an eye exam was performed. Patients who were lost to follow-up (LTFU) were defined as any subjects who did not receive any follow-up visits after their traumatic iritis diagnosis. The medical records of traumatic iritis patients were longitudinally reviewed through August 1, 2020.
Age- and sex-specific population figures for Olmsted County census data from 2006 to 2015 were used to calculate overall incidence and p-values. Age was stratified into the following age 0–14, 15–24, 25–44, 45–64, and ≥65 years. Incidence rates were also age- and sex-adjusted to 2020 census figures for the U.S. White population to make data comparable to national estimates as the Olmsted County population was approximately 83% White. A Poisson error distribution was assumed to calculate the 95% confidence intervals (95% CI) for incidence rates. Statistical analysis was performed using jamovi ver. 2.3.28.0 (2023), and SAS version 9.4 (Cary, NC).
There were 156 patients with an incident diagnosis of traumatic iritis during the 10-year study period. Females accounted for 35 (22.4%) patients (Table 1). The average age at traumatic iritis diagnosis was 33.4 years old (range: 4 to 96 years). In terms of race, 102 (65.4%) subjects were White, 24 (15.4%) were Black, 9 (5.8%) were Asian, and 8 (5.1%) were Hispanic. Compared to the mid-study (2010) population of Olmsted County, Minnesota, traumatic iritis was more frequently observed in male (p<0.001) and Black (p<0.001) patients. Medical comorbidities included hypertension in 22 (14.1%) patients, obesity in 6 (3.8%), smoking history in 46 (29.5%), of which 32 (20.5%) were active smokers at diagnosis, and autoimmune disease in 8 (5.1%). There were 4 (2.6%) patients with history of glaucoma, with 3 (1.9%) on IOP-reducing drops prior to traumatic iritis.
The overall incidence rate for all patients with traumatic iritis was 10.7 per 100,000 per year (Table 2). For men, incidence rate was 16.6 per 100,000 per year, while the incidence rate for women was 4.7 per 100,000 per year. Incidence was highest in the 15–24 age group with a rate of 29.1 per 100,000 per year, followed by the 25–44 age group at a rate of 22.6 per 100,000 per year. By mechanism, incidence was highest in sport-related injuries with a rate of 3.3 per 100,000 per year.
The most common mechanisms for traumatic iritis were sports-related (N=29, 18.6%), assault-related (N=23, 14.7%), scratch injury (N=22, 14.1%), and work-related injuries (N=21, 13.5%) (Table 3). Within the sport-related mechanisms (N=29, 18.6%), racquet sports (N=8, 5.1%), basketball (N=6, 3.8%), baseball (N=5, 3.2%), and soccer (N=5, 3.2%) were the most frequent. Within assault related mechanisms (N=23, 14.7%), punch injuries (N=16, 10.3%) were most frequent. Within scratch injuries, finger pokes (N=11, 7.1%) and tree branches (N=8, 5.1%) were most common. Within work-related mechanisms (N=21, 13.5%), metal foreign body (N=14, 9.0%) was most common. A large subset of patients suffered traumatic iritis due to projectile injuries (N=17, 10.9%), with thrown objects (N=6, 3.8%) and gun-related injuries (N=4, 2.6%) most frequent in this group. Other mechanisms for traumatic iritis included garden work (N=7, 4.5%), elastic cord snap, (N=7, 4.5%), falls (N=6, 3.8%), and motor vehicle accidents (N=5, 3.2%).
The mean time from traumatic event to iritis diagnosis was 3 days (range: 0 to 61 days). The most frequently reported presenting symptoms were eye pain (N=126, 80.8%), photophobia (N=100, 64.1%), and decreased vision (N=70, 44.9%) (Table 4). Tearing (N=22, 14.1%) and floaters (N=10, 6.4%) were less common presenting symptoms. In addition, there were 5 (3.2%) patients with binocular diplopia due to comorbid orbital fracture, 2 (1.3%) patients with monocular diplopia, and 1 (0.6%) with unspecified diplopia.
The mean initial Snellen best-corrected visual acuity (BCVA) of the affected eye at initial presentation was 20/40. Mean intraocular pressure (IOP) of the affected eye at presentation was 13.3 mmHg (range: 4 to 26 mmHg). The average cell grade in 140 eyes was +1.5 (range: +0.5 to +4), and the average flare grade in 55 eyes was +0.9 (range: +0.5 to +4). Besides anterior chamber cell and flare, the most common exam findings at initial presentation included conjunctival injection (N=108, 69.6%), corneal abrasion (N=23, 14.7%), and commotio retinae (N=22, 14.1%). Posterior synechiae was noted in 3 (1.9%) patients, traumatic cataract in 2 (1.2%) patients, and Vossius ring in 0 patients. There were 6 (3.8%) patients documented to have retinal hemorrhage, 1 (0.6%) patient with a retinal tear, and 0 patients with retinal dialysis. Of note, there were 7 (4.5%) of patients where findings of dilated fundus exam were not documented due to patient being unable to cooperate with exam (N=3, 1.9%) or omission by the examiner (N=4, 2.6%). Computed tomography of the head and/or orbits was obtained in 19 (12.2%) patients, of which a comorbid orbital fracture was noted in 8 (5.1%) patients. Microhyphema or hyphema was observed in 16 (10.3%) patients and resolved an average of 22.4 days (range: 4 to 72 days) after diagnosis. No patients with hyphema experienced re-bleeding or IOP spikes, and surgical intervention for anterior chamber washout was not required in any cases. There were 14 (9.0%) patients who had gonioscopy documented on examination, of which 3 (1.9%) patients were found to have angle recession on gonioscopy, none of whom were found to have cyclodialysis nor traumatic glaucoma in subsequent follow-up visits, with final ophthalmology follow-up at a mean of 7.0 years. The mean IOP at iritis-specific final follow-up was 14.8 mmHg (range: 8 to 24 mmHg) and mean Snellen visual acuity of the affected eye at iritis-specific final follow-up was 20/30 (range 20/20, 20/400), with the patient of 20/400 vision having a comorbid vitreous hemorrhage.
Topical corticosteroids (N=127, 81.4%) and topical cycloplegics (N=119, 76.3%) were the most frequently used treatment modality for traumatic iritis (Table 4). There were 97 (62.2%) patients who were given both cycloplegics and corticosteroids, 30 (19.2%) who received corticosteroids without cycloplegics, 22 (14.1%) who received cycloplegics without steroids, and 7 (4.5%) who were given neither. Out of 127 patients who were treated with topical steroids, 5 (3.9%) patients developed steroid response, of which 2 (1.6%) received IOP lowering drops while the other 3 discontinued topical steroids. Topical antibiotics were also used frequently (N=58, 37.3%). Of the 29 (18.6%) patients who were not prescribed topical corticosteroids, 16 (10.3%) was due to concern for infection from corneal epithelial injury. The mean number of iritis-specific ophthalmology follow-up appointments was 2.1 (range: 0 to 26), and the median time from diagnosis to final iritis-specific follow-up appointment in days was 11 days (range: 1 day to 1.6 years) for patients with any follow-up, with one patient having a particularly long follow-up period (1.6 years) due to recurrent inflammation after initial event.
A comparison of demographic and clinical factors in patients who were lost to follow-up (LTFU) compared to patients who received follow-up was performed (Table 5). With regards to demographics factors, Black patients were associated with higher frequency of LTFU (p=0.001). A history of smoking was also associated with more LTFU patients (p=0.004). With regards to clinical factors, patients with exam findings of corneal abrasion (p=0.02) were more likely to follow-up.
Few studies have investigated the clinical characteristics of traumatic iritis, and a vast majority of traumatic iritis studies were single-center based. This population-based retrospective cohort study investigated the incidence and clinical characteristics of traumatic iritis in a predominantly White U.S. Midwestern population. In previous reports, Engelhard et al. investigated traumatic uveitis in a single tertiary care center from 1984 to 2014. In a cohort of 54 patients, their patient population was 70.4% male, mean age at diagnosis was 31.2 years, and most common treatment modality was topical steroids in 77.8% cases.^6^ Meanwhile, Rosenbaum found that in a cohort of 24 patients with traumatic iritis, their patient population was 79.4% male with a mean age of 31 years.^3^ In comparison, this study’s patient cohort was 77.6% male, mean age of diagnosis was 33.4 years, and topical corticosteroids was the most common treatment modality as it was given to 81.4% of patients, yielding similar results.
The present study confirms previous reports that most traumatic iritis patients tend to be young males.^3,6^ We found a statistically significant association to males when compared to mid-study population (p<0.001) and a striking difference in traumatic iritis incidence rates between males versus 16.6 versus 4.7 per 100,000 per year, respectively. Another important demographic finding was increased frequency of traumatic iritis in Black patients (p<0.001). Racial and ethnic disparities in ophthalmology were likely contributing factors to these findings as it has been well established that underrepresented patient populations are at higher risk for various eye disease and visual impairment.^12–14^ In addition, all patients except for one had unilateral iritis, similar to findings reported by Engelhard et al. and contrasting Rosenbaum and colleagues’ older findings that reported one-third of patients had bilateral inflammation.^3,6^
We explored mechanisms of traumatic iritis in our cohort, which was only previously investigated by Moreno-Mendoza and colleagues in an ophthalmologic center in Mexico City.^15^ In their analysis of 447 patients, the most common traumatic mechanisms were thrown objects in 129 (28.80%) cases and punch/kick injuries in 99 (22.14%). In comparison, 29 (18.6%) of mechanisms were from thrown objects in this study, calculated by combining 23 (14.7%) sports-related injuries from a ball with 6 (3.8%) projectile injuries from thrown objects. In addition, there were 23 (14.7%) assault-related injuries (e.g., punch/kick/elbow). In contrast, the most common categorization for traumatic iritis mechanisms in this study was sport-related (N=29, 18.6%) and assault-related (N=23, 14.7%), which suggests that these mechanisms are high-risk situations for traumatic iritis.
When exploring factors associated with LTFU, Black patients were more frequently LTFU (p=0.001). This finding underscores systemic racial disparities in ophthalmology, as decreased adherence and follow-up rates among Black patients have been reported in other ophthalmic conditions such as glaucoma.^16,17^ Another demographic feature associated with LFTU for traumatic iritis was history of smoking, which is consistent with prior literature suggesting smoking is a risk factor for LTFU in other domains of healthcare.^18^ Of note, there was a statistical difference between patients who received follow-up versus patients who were LTFU when examining days from trauma to diagnosis (p=0.04). However, given that the average times were similar (e.g., 3.4 days for patients who received follow-up versus 3.3 days for patients LTFU), this difference was not clinically relevant. Cell grade was another statistically significant factor between patients who received follow-up versus patients who were LTFU (p=0.04), but the difference in average cell grade was also not clinically significant (1.6+ for patients who received follow-up versus 1.2+ for patients lost to follow-up). Lastly, patients diagnosed with comorbid corneal abrasion were more likely to follow-up (*p=*0.016), perhaps because these patients had eye pain and/or decreased vision.
Limitations of this study included its retrospective design, which contributed to inherent variations in examination, evaluation, and follow-up. This was an especially important consideration when examining follow-up data, as patients with more severe disease or ocular comorbidities may have been more likely to follow up compared to those with mild disease. In addition, gonioscopy data were only recorded for 14 patients, with 3 having angle recession, so rates of angle recession and traumatic glaucoma may have been underreported. In addition, some traumatic cases may have been missed if patients received care outside of Olmsted County, but this is unlikely to impact findings given Olmsted County is geographically isolated from other care centers beyond Mayo Clinic and Olmsted Medical Group.^7^ However, cases exclusively managed by private optometry practices in Olmsted County are not captured by the Rochester Epidemiology Project, which may have led to our incidence rates being underreported. Although this study has limited generalizability of results due to Olmsted County’s lack of racial and ethnic diversity and relatively small population size, we still identified sociodemographic trends in traumatic iritis epidemiology.
In conclusion, this retrospective population-based cohort study investigated the incidence, mechanisms, and clinical characteristics of traumatic iritis in a U.S. Midwestern population over a 10-year period. The overall rate of traumatic iritis was 10.7 per 100,000 per year, with traumatic iritis patients most frequently observed in younger males. All but 1 case were unilateral. Black patients are disproportionately affected by traumatic iritis and more frequently LTFU. The most common categorizations of mechanism injury were sports-related, assault-related, and work-related. These findings may provide further evidence to support education and public health strategies (i.e., encouraging use of protective eyewear during at risk activities) to decrease the incidence of traumatic iritis and prevent vision loss.