Clinical characteristics and prognostic factors of posterior segment intraocular foreign body in a tertiary hospital

Background To identify the clinical characteristics, prognostic factors and visual outcomes in posterior segment IOFBs patients managed by PPV in a tertiary hospital. Methods A retrospective chart review was performed for 56 patients, who had PPV for IOFBs removal between November 2013 and November 2015. The mechanisms of injury, the nature of the IOFBs, the BCVA before and after the surgery, the penetrating site and the complications of the surgery were all collected. Univariate analyses were conducted to evaluate the prognostic factors. Results The mean age of the patients was 36.4 years. The nature of IOFBs was mainly metal. Most injuries were commonly caused by hammering the metal. The mean preoperative VA was 2.30 logMAR, and mean final VA was 0.92 logMAR. From univariate analysis, good visual outcome was correlated with the good visual acuity before surgery and poor visual outcome was correlated with the macular break and multiple surgeries. Conclusions In a tertiary hospital of eastern China, most cases of IOFBs were work-related. The prognosis of the patients was really well in the patients with good presenting visual acuity. Nevertheless the prognosis was not good for those patients who had macular injury or underwent several surgeries because of retinal detachment, epiretinal membrane or proliferative vitreous retinopathy. Good facilities for eye protection are urgently in demand for the workers indeed.


Background
Open-globe injury often can result in serious visual loss and afflict most patients in the developing country. IOFBs account for 18-41% of all open-globe injury [1,2]. Most post-traumatic IOFBs (58-88%) reside in the posterior segment [1][2][3][4]. The visual prognosis depends on the IOFB size, the zone of the injury, and the accessible treatment [4][5][6]. Basically IOFBs should be removed from sclera or sclerocorneal tunnel by PPV. Though more techniques have been applied to remove the IOFBs such as "Magnet Handshake" technique [7] and "Macula Protection by Perfluorocarbon Liquid" [8], the prognosis of the patients was not so good in some areas. China is the biggest developing country in the world. During the past twenty years, thousands of factories have sprung up in the east of China. Due to lack of the protection facilities, a lot of workers got hurt at work. Though the injury of the eye could be found immediately after the accident, not all patients sought the treatment timely. The visual outcomes could be totally different whether the patients would consult the doctor in time or not. With advancement in the microsurgical technique, those severely traumatized eyes can be saved from enucleation. However, some patients received several surgeries and still could not save their sight. The objectives of this study were to identify the clinical characteristics and prognostic factors in posterior segment IOFBs patients managed by 23-gauge PPV in the east of china.

Methods
The patients who were diagnosed with IOFBs at The Second Affiliated Hospital of Zhejiang University School of Medicine between November 2013 and November 2015 were included for this study. The follow-up ended up with the patients' last control. The study was approved by Institutional Review Board of Second Affiliated Hospital of Zhejiang University School of Medicine and conducted in compliance with guidelines of the Declaration of Helsinki. The complete history of the patients was taken at their first presentation. The initial BCVA of the patient was recorded using the Snellen chart. It was converted to a logMAR units for statistical analysis. An anterior segment ophthalmologic examination was performed with a slit-lamp (BQ 900; Haag-Streit, Berne,Switzerland). The location of the metallic IOFB was identified by a computerized tomography before operation. The collected data comprised age, gender, mechanism of injury, preoperative VA, initial ocular features, nature of foreign bodies, time interval between injury and IOFBs removal, postoperative VA and complications. All patients with a leaking wound underwent primary wound repair by general ophthalmologists before IOFBs removal surgery. Patients with a self-sealing wound underwent PPV surgery for the initial intervention. A classic three-port, 23-gauge vitrectomy technique was performed by three retinal specialists (J.M., Y.W. and L.Z.). The non-contact wide-angle vitreous surgery system was used during the PPV surgery. The corneal entry site was sutured with Nylon 10-0 and the scleral wound was repaired with Vicryl 8-0. If the view for performing the PPV surgery was obstructed by the traumatic cataract, a lensectomy or phacoemulsification procedure was also done during the surgery. The intraocular lens was not implanted for the first time. IOFBs were removed from the enlarged sclerotomy or limbal incision either with the intraocular forceps or the external magnet. The injured retinal areas, including retinal holes or detached retina were secured by endolaser photocoagulation, cryoretinopexy. Either gas or silicone oil was used for the intraocular tamponade. The patients were regularly controlled by the retinal experts of the Second Affiliated Hospital of Zhejiang University School of Medicine. A PMMA lens with iris fixation was implanted in the anterior chamber or a single-piece PMMA lens was implanted into the ciliary sulcus depending on the integrity of the lens capsule after 3 months. In addition, vitreoretinal operations were performed in case of complications, such as retinal detachment or macular pucker. A good visual outcome was defined as the final BCVA equal to or better than 20/40. A poor visual outcome was determined as final BCVA of less than 20/200.

Statistical analysis
Statistical analysis of data was performed using SPSS 17.0 (SPSS Inc., Chicago, IL). Continuous data were reported as mean ± standard deviation, and categorical data were reported as n (%). Paired sample t-test and chi-square test were used for comparing the preoperative BCVA to the postoperative BCVA. The predictive factors for visual outcomes were studied using univariate analysis (Fisher exact test, or Mann-Whitney U-test). A P value of 0.05 was considered statistically significant in this study.
For good visual outcome, univariate analysis showed that good presenting VA (Fisher exact test, P = 0.005) was the significant associated predictors (Table 6). There was no statistical significance for the patients with different entry site, with lens injury, retinal injury and endophthalmitis or not.
For poor visual outcome, univariate analysis showed that presence of macular break and undergoing multiple vitrectomy (Fisher exact test, P = 0.006, 0.0001) were the significant associated predictors (Table 7). No statistical significance was seen from the patients with different initial VA and entry site, with lens injury, retinal injury and endophthalmitis or not.

Discussion
In the developing country, IOFBs is a serious problem in a young working age population. In accordance with previous reports [5,[9][10][11], our study showed that the majority of patients (96.4%) were male, with a mean age of 40.8 years. This study found that 91.1% of the patients had work-related injury. Metal was the most common nature of IOFBs, accounting for 96.4% of the patients, which is similar to the other studies [4,12,13]. This study also revealed that the hammering the metal was the most common mechanism of injury (58.9%), same as the other study in china [14]. It is different from in Thailand, where the electric grass trimmer was the most common mechanism of injury [15].
There are several tools to remove the IOFBs, including external magnets, intraocular magnets and foreign body claw. Pars plana vitrectomy can be the first choice for the patients with the IOFBs in the posterior segment, though magnetic suction from the sclera is still used in the non-developed areas. With advancement of the surgical facilities and the techniques, 23-gauge vitrectomy was commonly recommended for the posterior IOFBs on account of the less damages and rapid recovery [16]. Sometimes 25-gauge vitrectomy was also used in some cases. However, 25-gauge PPV is not necessary for the patient with large IOFBs. The IOFBs are usually extracted from the corneoscleral limbus or the sclera. It    depends on the size of the IOFBs and integrity of the lens. Usually the IOFBs will be removed from the corneoscleral limbus if their diameters are more than 6 mm. For the small ones, the IOFBs can be taken out from the limbus or sclera. We would like to choose the limbus if the patients were diagnosed with traumatic cataract. We found that good presenting VA before surgery was a significant associated predictor for the good visual outcome. This is the same as the previous reports [10,17]. But the other study claimed that the initial BCVA was not the best reliable predictive factor for the final BCVA by the multiple correspondence analyses [18].
Poor presenting VA has previously been reported as an important predictive factor for poor visual outcomes [18]. However, our study showed that most patients (35/ 48) with the initial VA less than 20/200 had better final VA (better than 20/200). It is reasonable that the patients had low presenting VA if the patients suffered traumatic cataract and vitreous hemorrhage. The VA can be improved greatly with cataract extraction and vitreous hemorrhages removal if the macular of the patients remained integrity. Unfortunately the patients had low final VA when they had macular break. Others also reported that the most relevant parameters for a low final BCVA were the presence of a macular lesion [18], RD at presentation and large foreign body [19]. The prognosis was also not good for those patients had several surgeries because of retinal detachment, epiretinal membrane or proliferative vitreous retinopathy. It is due to delayed return visit or a long wait-list for the surgery.
Previous study showed that presence of endophthalmitis, relative afferent pupillary defect (RAPD) and initial RD was the significant associated predictors for the poor final VA [15]. Our study showed the different results. Among 12 patients diagnosed with endophthalmitis, only 3 patients had the low final VA less than 20/200. There is no statistical difference between the final VA less than 20/200 group and the final VA better than 20/ 200 group. RAPD was not included for this study since some patients had anterior chamber hemorrhages and iris injuries. Initial RD was not the significant predictive factor for the poor final VA in our study. If the detached retina was reattached before the macular was involved, the final VA could be better. Recent study revealed that early removal of IOFB may related to the favourable visual outcome and low endophthalmitis [20]. It means the patient could have a higher chance to recover better if they got the timely treatment.

Conclusions
In conclusion, most cases of IOFBs were work-related in a tertiary hospital located in the east of china. The prognosis of the patients was really good with good presenting visual acuity. Nevertheless the prognosis was not good for those patients who had macular injury or underwent several surgeries because of retinal detachment, epiretinal membrane or proliferative vitreous retinopathy. Good facilities for eye protection are urgently in demand for the workers indeed.

Availability of data and materials
The datasets used and analyzed during the current study are available from corresponding author on reasonable request.
Author's contributions Authors' contributions: JM, YW and LZ performed the surgery, analyzed the patient data and made major contributions for writing the manuscript. MC and JA performed the literature review for similar topics and made major contributions to acquisition and interpretation of data. XY-F made substantial contributions to conception and design this study. All authors have read and approved the final manuscript.
Ethics approval and consent to participate This study was approved by the research ethics committee of The Second Affiliated Hospital of Zhejiang University School of Medicine. Due to the retrospective nature of this study, the inform consent from individual patient was waived by IRB in this study.