Clinical analysis of posterior segment injury in left-behind children in China

A retrospective review of children (aged< 16) with posterior segment injury presenting to the Department of vitreous and retinal between Jan 2012 and Jan 2015 was performed. This study was approved by the Institutional Review Board. Cases that underwent 25G minimally invasive vitrectomy were identified and patients with follow-up of 18 months or more were included in this study.

Data included the patient’s baseline character (gender, age), eye condition (visual acuity, trauma type, retinal detachment and intraocular infection etc.) and surgical methods (whether lens removal, intraocular lens (IOL) implantation, intraocular tamponade, or enucleation etc.). we divided visual acuities into 5 grades: No light perception (NLP), light perception/hand move (LP/HM), count finger to 0.1 (CF to 20/200), 20/200 to 20/50, and ≥ 20/40.

According to previous studies [8, 9], LBC were defined as children who were under 18 years old and were left at home with both or one of their parents migrate to urban areas for at least 6 months. Children with one of their patients leaving away for less than 6 months were defined NLBC.

Severity of ocular trauma of all patients were evaluated using the ocular trauma score (OTS) [10], and the OTS score was calculated according to the factors affecting vision. OTS-1 (0–44 points); OTS-2 (45–65 points); OTS-3 (66–80); OTS-4 (81–91); OTS-5 (92–100).

25G-PPVs were performed in all children. In cases with traumatic cataract or rupture of the capsule, lens was removed using phacoemulsification. At the end of surgery, long effect gas C3F8 (perfluorocarbon) or silicone oil was used for intraocular tamponade, depending on the damage of the retina. For cases with endophthalmitis, ceftazidime (2.25 mg/0.1 ml) and norvancomycin (1.0 mg/0.1 ml) were given. Surgical indications included: 1) I, II area trauma accompanied with vitreous, and retina traction; 2) III area trauma accompanied with retinal incarcerated or retinal traction; 3) traumatic retinal detachment, including serrated and tractional /rhegmatogenous retinal detachment, and complex retinal detachment with retinal incarceration; 4) vitreous hemorrhage after trauma obscures the macula, which is not absorbed after a month [11].

A total of 92 cases (92 eyes) at the Affiliated Ophthalmology and Otolaryngology Hospital of Fudan University underwent vitreous surgery treatment, including left-behind children (n = 48) and guarded children (n = 44), 50 cases were male, 42 cases were female, as show in Table 1. Age range 2–16 years old, mean patient age was 8.5 ± 4.6 years. Twenty-seven (61.4%) patients had open globe injuries, four (9.1%) patients had intraocular foreign body, six (13.6%) patients had closed globe injury, all from the guarded group. The follow-up time ranged from 18 to 32 months, and no one was lost. LBC had a higher incidence of infective endophthalmitis than NLBC. LBC are more likely to develop Relative Afferent Pupillary Defect (RAPD) than NLBC, and OTS rating is worse than NLBC. Totally two cases in LBC group and one case in NLBC group was excluded. One case in LBC group had been operated before using 20G PPV (sterilized air tamponade) in another hospital. The other case in LBC group could not afford the travelling expenses and was advised to be followed-up in the hospital in the vicinity of his home. The excluded case in NLBC group was a 12 years-old boy and had intraocular foreign body (IOFB) without lens or retinal damage. The child received simple 25G-PPV and IOFB removal, and his visual acuity recovered to 20/40 one week after surgery. He was followed up for a total of 7 months until he was sent abroad for junior high school. These three children were followed up by telephone and never come back to our department, therefore they were excluded from our study.

The comparison of the initial and final visual acuity between LBC and NLBC is shown in Table 2. In terms of composition ratio, the proportion of the initial and final eyesight of LBC above the index is significantly smaller than that of the NLBC. The visual acuity of LBC was elevated in 18 eyes (37.5%), 21 eyes (43.75%) without changes, and 9 eyes (18.8%) were decreased. The visual acuity of NLBC was elevated in 29 eyes (65.9%), 9 eyes (20.5%) without changes, and 6 eyes (13.6%) were decreased. A IOFB was removed from the left eye of the patient (Fig. 1.).

The days from the occurrence of trauma to the hospital admission in LBC group were 9.3 ± 5.7 days, and that in the guarded group were 4.4 ± 2.8 days (Fig. 2.). The length of days from the onset of trauma to the first 25G-PPV in LBC were 35.3 ± 25.1 days, and that in NLBC were 14.3 ± 7.7 days. The length of medical treatment of LBC was significantly longer than that of NLBC.

Vitrectomy was performed on average 2.1 ± 1.0 times, with LBC 2.3 ± 1.0 times and NLBC 1.9 ± 1.0 times. The comparison of vitreous surgery and long-term maintenance status between LBC and NLBC is shown in Table 3. As can be seen from the table, LBC had significantly higher proportion of intraocular lens resection, silicone oil tamponade and endophthalmitis than NLBC. The proportion of LBC who ultimately retain the lens or have the condition to be implanted with IOL was lower than that of NLBC, while the proportion of silicone oil dependence in LBC was higher. In LBC group, there was one case of secondary glaucoma and one case of eyeball enucleation, which were not found in NLBC group.