Ocular injuries are the leading cause of monocular visual disability and monocular blindness in children.1 2 Open-globe injuries are among the most serious types of ocular trauma, and up to 43% of these injuries are reported to occur in individuals less than 18 years old.3 Preoperative evaluation of children is often hindered by inadequate history, poor patient cooperation and assessment of visual outcome during the physical examination. In addition, follow-up is often short, particularly in eyes with less severe injuries. Complex open-globe injuries lead to vitreous haemorrhage causing tractional retinal detachment and occlusion amblyopia.4

Kuhn and associates5 described the ocular trauma score (OTS), a simplified categorical system for standardised assessment and visual prognosis in ocular injuries, in a large series including both paediatric and elderly patients.

The purpose of this study was to analyse the prognostic value of OTS in children with open-globe injuries.

METHODS

We conducted a retrospective review of 54 paediatric patients (age ⩽16) who presented to the Beyoglu Eye Research and Training Hospital between January 2001 and October 2006. Age, gender, type of injury and past ocular history were determined from each patient’s medical record. Thirty of 54 consecutive children presenting with open-globe injury were included in the statistical analysis. We accepted only patients over 2 years of age for whom initial visual acuity for OTS variables (table 1) were available. Twenty-four patients (24 eyes) were excluded where complete data for OTS variables (type of injury, afferent pupillary defect) were not available. Children under 2 years of age were excluded because an accurate assessment of their presenting visual acuity data was considered unreliable.

All patients had experienced only ocular trauma without accompanying body trauma, and none of the patients in our series had suffered bilateral injuries.

Patching treatment for amblyopia was started as soon as possible after the eye was repaired. Refractive abnormalities such as regular astigmatism were treated with spectacle lenses. Patients with irregular astigmatism and those who were aphakic were treated with contact lenses.

The type of injury was classified according to Birmingham Eye Trauma Terminology.6 A numerical value (raw number) was assigned to each of the following variables: A• initial visual acuity, B• rupture, C• endophthalmitis, D• perforating injury, E• retinal detachment, F• afferent pupillary defect. The OTS was determined by calculating the sum of the raw points: A+B+C+D+E+F. If a specific variable was not present, its value was zero. Similarly to the OTS study group, presenting visual acuities were divided into five categories (1 to 5) as 20/40⩽, 20/200–20/50, 1/200–19/200, LP/HM and NLP in our study (table 1). Certain numerical values render to OTS variables were converted to OTS categories. The lowest possible OTS score is zero, whereas the highest is 100.

A categorical distribution of 30 eyes in our study and 2151 eyes in the OTS study was carried out. The percentages of eyes in each category of the OTS study and our study were compared. The chi-square and Fischer exact tests were used for statistical analysis. A p value of less than 0.05 was considered statistically significant.

RESULTS

The mean age of 30 patients was 7 (SD 0.25) (2–16) years. The mean ages in each category were 5 (SD 0.05) (2–12) years in category 1, 6 (SD 0.14) (2–14) years in category 2, 7 (SD 0.62) (5–16) years in category 3, 7 (SD 0.62) (5–16) years in category 4, and 12 (SD 0.04) (7–16) years in category 5. The male/female ratio was 12 (40%) to 18 (60%). Right and left eyes were involved in 75% and 25% of patients, respectively. The mean follow-up was 18.2 (4.1) (6–40) months.

The percentage of final visual acuity in each OTS categories in our study group is compared with the OTS study group, and the results are demonstrated in table 2. In table 2, “A” represents the percentage of eyes in OTS study group, whereas “A′” represents the percentage of eyes in our study group in each category.

When we compared the table of distribution of final visual acuities in all OTS categories, NLP and 20/200 to 20/50 ratios in the OTS study were significantly higher than in our study (p<0.001), whereas LP/HM ratios were significantly higher in our group than in the OTS study (p<0.001).

When we compared final visual acuities in each categories of our study and the OTS study, in our group, no light perception (NLP) ratios were lower in the first two categories (p = 0.012, p<0.001, respectively); light perception (LP)/hand motion (HM) ratios, and 1/100–19/100 ratios were higher in category 2 and in category 3 (p<0.001 each), and the 20/200 to 20/50 ratio was lower in categories 3, 4 and 5 (p<0.001, p<0.001, p = 0.028, respectively). The ratio of ⩾20/40 was lower in categories 2 and 3 (p<0.001, each) but higher in category 4 (p<0.001) in our study.

DISCUSSION

Traumatic ocular injuries often cause significant visual loss in young patients, as well as in adults, and are the leading cause of non-congenital unilateral blindness in children.1 7

Prediction of visual outcome has major significance for the patient, patient’s family (anxiety relief, quality of life issues, economical decisions) and patient’s ophthalmologist (counselling, triaging, management, rehabilitation). The OTS uses a limited number of variables (readily determined at the time of the initial examination or surgery) and basic mathematics to enable the ophthalmologist to predict the final functional outcome up to 77% within ±one visual category shortly after the eye injury.

Comparing the general distribution of final visual acuities between our study and OTS study, the OTS study results were found to be clustered in NLP and 20/200–20/50 range, whereas the majority of our findings were in the LP/HM range. Evaluating subgroups in each category, the final visual acuity increased proportionally as the number of the category increased in the OTS study. However, in our study, we observed a proportional increase in final visual acuity in categories 4 and 5 only, and no proportional increase was detected in the first three categories. The mean age of patients in our study and especially the mean ages of patients in the first four categories are in the range associated with amblyopia. This may have played a significant role in the poor visual outcome of children with open globe injuries along with the potential damages of trauma on the ocular structures. In addition, children may develop more extensive postoperative inflammation, scarring, and proliferative vitreoretinopaty (PVR) than adults, which may also affect the anatomical and functional outcomes.10^–12

A problem specific to children with ocular trauma is the development of amblyopia in the injured eye. The younger the child at the time of visual deprivation, the more rapid the development of amblyopia.8 9 Children under the age of 8 years are at particular risk for amblyopia due to open-globe injuries. Therefore, vigorous anti-amblyogenic treatment should be performed for this paediatric age group.

Paediatric age-related factors (such as inadequate history, poor patient cooperation, communication difficulties at the initial examination, consecutive occlusion amblyopia and extensive postoperative inflammation/scarring) might also limit the accurate prediction of visual outcomes with OTS in this population.

Sobacı et al13 showed that deadly weapon-related open-globe injuries, especially those associated with land mines and hand grenades, had devastating visual consequences. They found that the final visual acuities in OTS categories in their group were similar to those in the OTS study group, except for LP/HM in category 2 (53% vs 26%), and their study group found no patients in category 5.13 In our series, no patient had a terror-related injury, and none of the patients in our study had associated body trauma or systemic problems.

Our results may have been affected by the fact that ours is a retrospective, non-randomised, small case series with possible treatment selection bias.

In conclusion, keeping these limitations in mind, the OTS may provide a gross prediction of final visual acuity in paediatric patients with open-globe injuries. Further prospective studies with larger series and with longer follow-up times are warranted.