Despite growing prevention efforts, ocular injuries remain a major cause of blindness with an estimated incidence of 2.4 million cases in the United States [
The aim of this study is to identify predictive factors for visual outcomes of patients presenting with a posterior segment IOFB treated in a tertiary university hospital in Quebec, Canada, and to review the surrounding literature.
This retrospective study was approved by the Institutional Review Board of the Centre Hospitalier Universitaire de Quebec–Université Laval and adheres to the tenets of the Declaration of Helsinki.
All consecutive patients treated, hospitalized, or operated for posterior segment IOFB at the CHU de Quebec–Université Laval, a university tertiary hospital, between January 1, 2009, and December 31, 2018, were included in the study. Exclusion criteria included nonposterior segment IOFB and patients who had immediate exenteration, enucleation, or evisceration.
Data collection included age, gender, VA preoperatively and at least one month after the last intervention, final diagnosis, IOFB characteristics (localization, size, nature, and entry site), and delay between the injury and the surgery in days. The type and date of interventions, extraction type (pars plana or anterior chamber), surgical procedure (vitrectomy, buckle, endolaser photocoagulation, type of tamponade, use of a corneal graft, and phacoemulsification (none, first or second intention)), intraocular lens implantation (none, primary or secondary lens implantation), the number and the type of additional diagnoses and complications at presentation (e.g., endophthalmitis, cataract, hyphema, vitreous hemorrhage, retinal tear, retinal detachment, iris damage, choroidal detachment, or hemorrhage), the need for a second intervention (excluding phacoemulsification and intraocular lens implantation (phaco-IOL)), and an evisceration or enucleation were also collected. The context and mechanism of the injury as well as the presence of a relative afferent pupillary defect (RAPD) were not collected because of incomplete data related to these variables.
The surgical technique used to remove IOFB depended on clinical characteristics but usually involved a pars plana vitrectomy with IOFB extraction using forceps or magnets, followed by the use of a tamponade. Primary repair also varied depending on clinical characteristics but was generally performed using permanent suture of varied size depending on the localization of the entry wound.
Data are presented as mean ± standard deviation (SD) for continuous variables and as frequencies (percentages) for categorical variables. Best corrected visual acuity (BCVA) was converted to an Early Treatment Diabetic Retinopathy Study (ETDRS) letter score from the metric Snellen scale for analysis. For the purposes of analyses, missing data for delay before presentation was imputed using the median value for the 5 (9%) missing cases. Patients were separated into two groups based on final VA: the good VA group with vision greater than 50 ETDRS letters (Snellen equivalent: 20/100) and the poor VA group with vision less than or equal to 50 ETDRS letters. Comparisons of both groups were performed to identify characteristics associated with a poor final VA. The following prognostic factors were analyzed for possible associations with final VA: age, time of injury to intervention, entry site, IOFB characteristics (i.e., location, size, nature), extraction site, clinical features on initial presentation (number of preoperative additional diagnoses, the presence of cataract, retinal tear, retinal detachment, vitreous hemorrhage, hyphema, iris damage, choroid detachment, endophthalmitis), IOL implantation (none, first or secondary implantation), the need for a secondary intervention (excluding phaco-IOL), and the need for a scleral buckle. Any damage to the iris that could reasonably be associated with the trauma was considered as an iris damage, regardless of size or severity. Characteristics and variables were compared between the two groups using independent Student’s
A multiple logistic regression model was built to identify risk factors for worse final VA. A backwards elimination strategy was used to select variables with variables
A total of 54 patients were included in the study (Table
Clinical characteristics of 54 study participants.
Clinical characteristics | Mean ± SD or |
---|---|
IOFB size, mm | 5.31 ± 4.62 |
IOFB nature | |
Metallic | 48 (89%) |
Glass | 3 (6%) |
Fireworks | 1 (2%) |
Organic | 1 (2%) |
Unknown | 1 (2%) |
IOFB entry site | |
Cornea | 30 (56%) |
Sclera | 15 (28%) |
Corneoscleral | 8 (15%) |
Unknown | 1 (2%) |
IOFB location | |
Retina | 31 (57%) |
Vitreous | 19 (35%) |
Orbit | 2 (4%) |
Optic nerve | 1 (2%) |
Unknown | 1 (2%) |
Secondary clinical features | |
Cataract | 39 (72%) |
Vitreous hemorrhage | 28 (52%) |
Hyphema | 17 (31%) |
Retinal detachment | 16 (30%) |
Retinal tears | 8 (15%) |
Iris damage | 6 (11%) |
Choroid detachment | 3 (6%) |
Endophthalmitis | 1 (2%) |
IOFB, intraocular foreign body; SD, standard deviation.
Histogram illustrating number of intraocular foreign bodies per year during the study period.
The most common additional diagnoses and complications seen with IOFB were cataracts (
Thirty-three percent of patients presented with an initial VA of 6/18 or better and 15% of patients had an initial VA of 6/9 or better. Forty-four percent of patients presented with an initial VA of no light perception (NLP), light perception (LP), or hand motion (HM). On final presentation, 50% of patients had a VA of 6/18 or better and 33% of patients had a final VA of 6/9 or better. Moreover, only 19% of patients had a final VA of NLP, LP, or HM (Table
Initial and final VA.
VA | Initial VA | Final VA |
---|---|---|
NLP, LP or HM | 24 (44%) | 10 (19%) |
CF to 6/60 | 8 (15%) | 16 (30%) |
6/45 to 6/21 | 4 (7%) | 1 (2%) |
6/18 to 6/12 | 10 (19%) | 9 (17%) |
6/9 to 6/6 | 8 (15%) | 18 (33%) |
CF, counting fingers; HM, hand motion; LP, light perception; NLP, no light perception; VA, visual acuity.
Among the 54 patients, 21 (39%) received surgical intervention within 24 hours of the incident. IOFB removal was achieved via the pars plana in 48 cases (89%) or via the anterior chamber in 5 cases (9%). All patients underwent a vitrectomy. Nineteen patients (35%) had a scleral buckle (combined to the vitrectomy) and 5 (9%) had a corneal graft. 43 patients (80%) underwent phacoemulsification. Thirty-two eyes (59%) had phacoemulsification with secondary intraocular lens (IOL) implantation while only 20% (
The results revealed that poor final BCVA was associated with the following factors: a poor baseline VA (
Baseline characteristics of 54 patients with intraocular foreign body by final visual acuity.
Characteristics | Final visual outcome | ||
---|---|---|---|
Good >50 letters ETDRS | Poor ≤50 letters ETDRS | ||
Mean ± SD or | |||
Male sex | 28 (93%) | 22 (92%) | 0.82 |
Age | 38.5 ± 12.5 | 43.5 ± 12.1 | 0.15 |
Baseline VA | 57.9 ± 30.9 | 12.5 ± 24.3 | |
Delay | 20.0 ± 66.7 | 9.8 ± 39.3 | 0.47 |
IOFB | |||
Metallic nature | 28 (93%) | 19 (83%) | 0.22 |
Size | 2.8 ± 1.7 | 7.5 ± 5.2 | |
Localization | |||
Vitreous | 13 (43%) | 6 (25%) | 0.16 |
Retina | 17 (57%) | 14 (58%) | 0.90 |
Entry site | |||
Cornea | 17 (57%) | 13 (57%) | 0.99 |
Corneoscleral | 3 (10%) | 5 (22%) | 0.24 |
Sclera | 10 (33%) | 5 (22%) | 0.35 |
Number of additional diagnoses | 1.7 ± 1.0 | 3.5 ± 1.4 | |
Vitreous hemorrhage | 10 (33%) | 18 (75%) | |
Cataract | 20 (67%) | 19 (79%) | 0.31 |
Retinal detachment | 6 (20%) | 10 (42%) | 0.08 |
Retinal tear | 3 (10%) | 5 (21%) | 0.27 |
Hyphema | 3 (10%) | 14 (58%) | < |
Iris damage | 1 (3%) | 5 (21%) | |
Choroid detachment | 1 (3%) | 2 (8%) | 0.43 |
Endophthalmitis | 0 (0%) | 1 (4%) | 0.26 |
Interventions | |||
Anterior chamber extraction | 0 (0%) | 5 (21%) | |
Pars plana extraction | 30 (100%) | 24 (100%) | 1.00 |
Scleral buckle | 8 (27%) | 11 (46%) | 0.14 |
Graft | 2 (7%) | 3 (13%) | 0.46 |
First intention IOL | 8 (27%) | 3 (13%) | |
Reoperation | 4 (13%) | 10 (42%) | |
Tamponade | |||
Air | 17 (57%) | 6 (25%) | |
SF6 | 11 (37%) | 4 (17%) | 0.10 |
C3F8 | 0 (0%) | 7 (29%) | |
Silicone | 2 (7%) | 7 (29%) | |
Postoperative VA | 79.6 ± 7.0 | 13.1 ± 14.9 | < |
Bolded figures = statistically significant results at the 0.05 level. Abbreviations: ETDRS, Early Treatment Diabetic Retinopathy Study; IOFB, intraocular foreign body; SD, standard deviation; VA, visual acuity.
For further clarification of independent risk factors for a poor visual outcome, a multiple logistic regression model was built (Table
Multiple logistic regression for predictive factors of worse final visual acuity following intraocular foreign body extraction in 54 patients.
Characteristics | OR (95% CI); |
---|---|
Age, years | 1.00 (0.93, 1.07); 0.97 |
Baseline VA, ETDRS | 0.97 (0.94, 0.99); |
Number of additional diagnoses | 2.95 (1.40, 7.56); |
Bolded figures = statistically significant results at the 0.05 level. Abbreviations: CI, confidence interval; ETDRS, Early Treatment Diabetic Retinopathy Study; OR, odds ratio; VA, visual acuity.
Despite growing awareness for the use of protective eyewear, IOFBs continue to be a common cause of blindness in both developed and undeveloped countries [
As in previous studies, patients are predominantly male with an average age of 40 years [
The majority (88%) of IOFBs in our study were inorganic and metallic in origin. In general, inorganic, metal IOFBs are better tolerated than organic matter. Most metals are inert and, if small and deeply lodged, may be left in place [
We also found a significant difference in IOFB size based on final visual prognosis (final VA > 50 letters = 2.8 ± 1.7 mm vs. final VA ≤50 letters = 7.5 ± 5.2 mm,
There was no significant association between entry site and visual prognosis. In contrast, an earlier study reported better visual prognosis with a scleral or corneoscleral entry site than a corneal site [
Complications identified as poor prognostic factors in our study were consistent with findings in the literature, including vitreous hemorrhage [
There are many conflicting studies regarding the management of IOFB in terms of the optimal time of removal. Recent studies suggest that immediate IOFB removal may not be as critical to vision preservation as previously thought [
Our study found no association between the time before removal and the final visual outcome. These results are consistent with Wickham et al. [
Only one-third of the patients in our study presented with an initial VA of 6/18 or better and 15% of patients had an initial VA of 6/9 or better. We found that better initial VA was associated with better final VA. This correlation has previously been found in numerous studies [
The requirement for repeated surgery is associated with a poorer visual outcome [
The use of C3F8 gas or silicone oil tamponades compared to air tamponade was associated with poor visual outcome in this study. Similar results were reported by Liu et al. with C3F8 [
This is a retrospective study with a relatively small sample size that provides epidemiological data for IOFB in our region. Our results are comparable to previous studies in other regions, including a study in North Carolina in which similar factors of poor visual outcomes (i.e., poor initial VA, the presence of a RAPD, and vitreous hemorrhage) were identified among a cohort of 59 predominantly young male patients [
In conclusion, this study identified multiple risk factors for poor VA following posterior segment IOFB over the span of a decade. Independent risk factors included worse baseline VA and a high number of concomitant complications (e.g., vitreous hemorrhage, hyphema, iris damage, and cataract). These factors will allow ophthalmologists to more accurately assess the trauma severity and visual prognosis of patients with a posterior segment IOFB. Future directions include assessing the impact on long-term visual acuity and the impact of the type of intervention on long-term visual prognosis.
All the data relevant to the study are included in the article. The datasets used and/or analysed during the present study are available from the corresponding author on reasonable request.
The authors declare that they have no competing interest related to the work submitted. Serge Bourgault and Eric Tourville are consultants and members of ad-board for Bayer and Novartis, outside the submitted work. Eric Tourville is also a consultant to Roche. Ali Dirani is a member of ad-board for Bayer and Novartis, outside the submitted work. Jean-Philippe Rozon, Mélanie Hébert, Eunice You, Guillaume Lavertu, and Mathieu Caissie have nothing to disclose.
The authors would like to thank the Centre Universitaire d’Ophtalmologie clinical research team for their support of this research study (Marcelle Giasson and Johanne Doucet).