Risk Factors for Endophthalmitis and Retinal Detachment with Retained Intraocular Foreign Bodies
Purpose. To analyze risk factors for endophthalmitis and retinal detachment (RD) in patients with retained intraocular foreign bodies (IOFBs). Design. A retrospective, interventional, consecutive case series. Participants. All patients treated at Bascom Palmer Eye Institute for traumatic IOFBs between 1999 and 2008. Methods. Analysis of visual outcome, mechanism of injury, management, and postoperative course. Results. 108 eyes with IOFBs were identified. Endophthalmitis occurred in 7 eyes (6.4%) at presentation, and risk was higher with vegetable matter exposure (P = 0.003). All eyes with posterior segment IOFBs received intravitreal antibiotics and there were no cases of endophthalmitis after initial management. RD was identified in 6 of 108 eyes (5.5%) at presentation. Risk factors were entry more than 5 mm behind the limbus (P < 0.001) and posterior segment IOFB (P = 0.028). Postoperative RD occurred in 11 of 102 eyes (10.7%). Risk factors for postoperative RD were preoperative endophthalmitis (P = 0.001), posterior segment IOFB (P = 0.008), and retinal impact sites (P = 0.028). Conclusions. Risk factors for endophthalmitis included vegetable matter exposure and delay to initial management. Risk factors for RD were posterior entry site, posterior segment IOFB, endophthalmitis, and retinal impact sites. No eyes developed endophthalmitis after presentation.
Hindawi Publishing Corporation
Journal of Ophthalmology
Volume 2012, Article ID 758526, 6 pages
Risk Factors for Endophthalmitis and Retinal Detachment with
Retained Intraocular Foreign Bodies
D. Wilkin Parke III, Avinash Pathengay, Harry W. Flynn Jr.,
Thomas Albini, and Stephen G. Schwartz
Department of Ophthalmology, Bascom Palmer Eye Institute, Miller School of Medicine, University of Miami,
Miami, FL 33136, USA
Correspondence should be addressed to Harry W. Flynn Jr., hﬂynn@med.miami.edu
Received 6 January 2012; Accepted 29 February 2012
Academic Editor: Christopher Leﬄer
Copyright © 2012 D. Wilkin Parke III et al. This is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly
Purpose. To analyze risk factors for endophthalmitis and retinal detachment (RD) in patients with retained intraocular foreign
bodies (IOFBs). Design. A retrospective, interventional, consecutive case series. Part icipants. All patients treated at Bascom Palmer
Eye Institute for traumatic IOFBs between 1999 and 2008. Methods. Analysis of visual outcome, mechanism of injury, management,
and postoperative course. Results. 108 eyes with IOFBs were identiﬁed. Endophthalmitis occurred in 7 eyes (6.4%) at presentation,
and risk was higher with vegetable matter exposure (P
= 0.003). All eyes with poster ior segment IOFBs received intravitreal
antibiotics and there were no cases of endophthalmitis after initial management. RD was identiﬁed in 6 of 108 eyes (5.5%) at
presentation. Risk factors were entry more than 5 mm behind the limbus (P<0.001) and posterior segment I OFB (P
Postoperative RD occurred in 11 of 102 eyes (10.7%). Risk factors for postoperative RD were preoperative endophthalmitis (P
0.001), posterior segment IOFB (P = 0.008), and retinal impact sites (P = 0.028). Conclusions. Risk factors for endophthalmitis
included vegetable matter exposure and delay to initial management. Risk factors for RD were posterior entry site, posterior
segment IOFB, endophthalmitis, and retinal impact sites. No eyes developed endophthalmitis after presentation.
Retained intraocular foreign bodies (IOFBs) are potential
occur in 10–41% of open-globe injuries [1–4]. Severe visual
loss may be associated with the original injur y, secondary
endophthalmitis, r hegmatogenous retinal detachment (RD),
or a variety of long-term complications including glaucoma,
cataract, inﬂammation, or foreign body toxicity. Traumatic
endophthalmitis and RD are of particular concern given
their tendency towards rapid progression resulting in severe
visual loss [3, 5–7]. Endophthalmitis has been reported in
2–30% of open-globe injuries with retained IOFBs [8–10],
and an IOFB may be present in 43% of eyes with traumatic
endophthalmitis . RD has been reported in 6–30% of
eyes with IOFBs [6, 12, 13].
Although the ideal timing of IOFB removal is early,
this may not always be clinically feasible. The current study
investigates risk factors for endophthalmitis and RD in
patients with retained IOFBs, employing a retrospective
cohort of patients with a history of IOFB treated at a single
ocular trauma center over a ten-year period.
2. Pa tients and Methods
The study was performed with the approval of the University
of Miami Institutional Review Board and in accordance with
the U.S. Health Insurance Portability and Accountability
Act and the Declaration of Helsinki. A database search was
performed via diagnostic and procedural codes of all patients
presenting to Bascom Palmer Eye Institute with open-globe
injuries and associated t raumatic IOFBs between January
1st, 1999 and December 31st, 2008. Surgically induced
retained IOFBs such as intraocular lenses were excluded
from the current study. Intraorbital, extraocular IOFBs were
excluded. IOFBs were deﬁned as intraocular if all or part of
2 Journal of Ophthalmology
the foreign material was within the intr aocular space at time
Medical records were reviewed for data regarding patient
demographics, mechanism and details of injury, visual acuity
(VA) and examination results at presentation and postoper-
atively, any diagnostic imaging, and surgery. The sensitivity
of the imaging modalities for recognition of IOFBs was
evaluated by identifying the IOFBs found at time of surgery
and searching for correct detection of these IOFBs on pre-
operative imaging reports. Surgical details included timing,
type, and number of procedures, a s well as the intraoperative
ﬁndings or administration of pharmaceutical agents. The
intervals between injury, presentation, and treatment were
noted. Postoperative followup was included extending up to
two years after injury when available. Factors directly con-
tributing to decreased visual acuity were determined when
possible from the retrospective review of the medical records.
The Ocular Trauma Classiﬁcation Group system for zone
of entry was applied, with corneal entry referred to as zone
one, entry within 5 mm posterior to the limbus as zone two,
and anything posterior to zone two as zone three . In
cases of large wounds crossing multiple zones, the zone in-
volved in the most posterior aspect of the entry wound was
used for this classiﬁcation.
Visual acuity was converted from Snellen to logMAR
scale for the purpose of calculating mean values and con-
verted back to Snellen score for data presentation. LogMAR
values of 2, 3, 3.5, and 4 were assigned to visual acuity scores
), hand motion (HM), light
perception (LP), and no light perception (NLP), respectively,
as have been utilized previously at the study institution
. Statistical analysis was performed with the two-tailed
student t-test for continuous variables and the Fischer test
for categorical data. P value less than 0.05 was considered
A total of 108 eyes in 107 patients had IOFBs and underwent
treatment; one patient had bilateral IOFBs. Mean age among
the 107 patients was 31.4 years (range 7–66), with 104 (97%)
male. The mean time from injury to presentation was 3.5
days with a range of 2 hours to 248 days. Excluding one
extreme value of 248 days, the mean time was 1.8 days with
range 2 hours to 18 days. Mean followup was 54.4 days, with
range 8 days to 2 years.
Mean best corrected visual acuity (BCVA) for al l eyes was
20/600 (logMAR 1.48, range 20/20 to NLP) preoperatively
and 20/210 (logMAR 1.03, range 20/20 to NLP) postopera-
= 0.038). Foreign body location within the eye (as
distinguished from zone of entry) was divided into anterior
segment, of which an intralenticular subset was identiﬁed,
posterior s egment (posterior to posterior lens capsule), and
traversing (long IOFBs protruding from the eye and extend-
ing into the posterior segment). Anterior segment IOFBs
were identiﬁed in 41 eyes (38.0%), of which 29 eyes (70.7%)
were anterior to the lens, with a mean preoperative VA of
20/80 (logMAR 0.60), and postoperative last corrected VA of
20/63 (logMAR 0.50) (P
= 0.84). Twelve of the 41 anterior
segment IOFBs (29.2%) were intralenticular, with preopera-
tive VA 20/370 (logMAR 1.275) and last corrected VA 20/36
(logMAR 0.26) (P
= 0.006). Fifty-seven eyes (53%) had pos-
terior segment IOFBs, with preoperative VA 20/2200 (log-
MAR 2.04) and last corrected VA 20/740 (logMAR 1.57) (P
0.122). Ten IOFBs (9%) were traversing, with preoperative
VA 20/520 (logMAR 1.42) and last corrected VA 20/350
(logMAR 1.25) (P
= 0.82). Six of 108 eyes (6%) underwent
enucleation, 2 as a primary surgery and 4 as a secondary pro-
Analysis of the IOFB type was available for 67 of 108
eyes (62%) (Table 1). The majority of foreign bodies were
magnetic. Fifty six of 67 samples (84%) were described on
pathology report as being magnetic, versus 2 (3%) that were
nonmagnetic metal, and 9 (13%) that were nonmetallic.
The mean largest dimension of the IOFB on pathology was
5.4 mm (range 0.5–17 mm). Sixty eight of 108 eyes (63%)
had IOFBs visible on clinical examination alone at presen-
tation, and 12 (11%) were protruding through the presumed
entry site at time of examination. The entry site was identi-
ﬁable preoperatively in 93 eyes (86%), intraoperatively in an
additional 4 eyes (4%), and could not be determined in 11
eyes (10%). Fifty two eyes (48%) had zone one, 38 eyes (35%)
had zone two, and 7 eyes (6%) had zone three entry. Zone
three entry was associated with a higher r ate of enucleation
(4 of 7 eyes, P<0.001) and poor ﬁnal visual acuity (<5/200
in the two nonenucleated e yes).
Time interval to surgical removal of retained IOFBs
was measured from presentation and from reported injury.
Two of the 108 eyes (2%) underwent primary enucleation.
One hundred and three of the remaining 106 eyes (97%)
underwent removal of IOFB within 24 hours of presentation.
This stands in distinction from the interval from injury
to IOFB removal, which was much more variable. Mean
time from injury to IOFB removal was 3.1 days, excluding
two outliers which were operated 137 and 276 days after
injury, both of which had occult posterior segment IOFBs
with moderate inﬂammation. Anterior segment IOFBs were
removed a mean of 3.3 days after injury, intralenticular
IOFBs a mean of 14.4 days after injury, posterior segment
IOFBs 3.1 days after injury, and traversing IOFBs 1.1 days
after injury. The 3 eyes (3%) in which IOFB removal was
delayed more than 24 hours after presentation underwent
IOFB removal at 7, 28, and 137 days after presentation.
Removal was delayed 7 days in the ﬁrst patient because the
patient left against medical advice prior to the ﬁrst scheduled
surgery and then returned 5 days later. Final BCVA in this
patient was 20/800. The second patient presented 248 days
after injury with an intralenticular IOFB and was observed
for 28 days prior to removal, and ﬁnal BCVA was 20/20. The
third patient had an intralenticular IOFB that was observed
for 137 days until visually signiﬁcant cataract developed, and
ﬁnal BCVA was 20/20.
Endophthalmitis occurred in 7 of 108 eyes (6.5%),
all clinically evident at time of presentation before initial
surgery (Ta b l e 2). Al l 7 were culture positive. No eyes devel-
oped endophthalmitis after IOFB removal during followup.
Five of the 7 eyes (71%) with endophthalmitis upon pres-
entation had posterior segm ent IOFBs, and 2 of 7 (29%)
Journal of Ophthalmology 3
Table 1: IOFB characteristics and visual acuit y by location.
No. of IOFBs
Mean zone of
All locations 108 56/67 (84) 3.93 1.27 20/600 20/210
29 (27%) 10/16 (63) 3.45 1.2 20/80 20/63
Intralenticular 12 (11%) 6/7 (86) 1.67 1.00 20/370 20/36
57 (53%) 30/34 (88) 3.78 1.35 20/2200 20/740
Traversing 10 (9%) 10/10 (100) 14.5 2.00 20/520 20/350
Table 2: Endophthalmitis in eyes with IOFBs: demographics, organisms, and outcomes.
Culture result IOFB
1 55 Mowing lawn 4 4
wire 16 mm
Bacillus sp., S.
6 39 Mowing lawn
wire 14 mm
NLP NLP Enucleation
had traversing IOFBs. Both of the traversing IOFBs causing
endophthalmitis were magnetic wires extending through
anterior segment, lens, vitreous, and into the retina. Preop-
erative VA in the eyes with endophthalmitis was HM in 6
eyes and NLP in 1 eye (mean logMAR 3.14). Postoperative
last corrected VA in the endophthalmitis eyes was 10/200 in
1 eye, 5/200 in 3 eyes, 1/200 in 1 eye, HM in 1 eye, and NLP in
1 eye (the same eye that was NLP preoperatively). The mean
logMAR visual acuity on last examination was 2.16. The eye
that was NLP after initial repair underwent secondary enu-
cleation. Risk factors for endophthalmitis included a mecha-
nism of injury related to vegetable matter exposure from yard
work (3 of 7 eyes, P
= 0.003), nonmetallic IOFBs (3 of 7 eyes,
= 0.081), and time interval from injury to presentation
(2.7 days versus 1.8 days in eyes without infection, P
All patients with IOFBs involving the posterior segment
or posterior capsular violation received prophylactic intrav-
itreal antibiotics (vancomycin 1 mg/0.1 cc and ceftazidime
2.25 mg/0.1 cc) at time of pr i mary surgery. Use of prophy-
lactic intravitreal or intracameral antibiotics with anterior
segment IOFBs occurred in 35 of 41 eyes (85.4%). Systemic
antibiotic usage varied over the ten-year study period, but
generally included a single dose of an intravenous ﬂuoro-
quinolone. No patients developed direct evidence of siderosis
or toxicity due to the foreign body substance during the
RD occurred in 18 eyes (16.7%). The presence of RD
before initial surgery and postoperatively was analyzed sep-
arately. The retina was visible on initial fundus examination
in 66 of 108 eyes (61.1%), not visible in 35 eyes (32.4%), and
not noted in the record in 7 eyes. Six out of 108 eyes (5.6%)
had an RD preoperatively. Risk factors for RD at presenta-
tion were zone three entry site and posterior segment IOFB.
Six of 6 eyes with RD at presentation had Zone Two or Three
injury, compared to 39 of the other 102 eyes (P
The mean value for zone of entry was 2.7 in eyes with pre-
operative RDs, as opposed to 1.3 in the eyes not presenting
with RD (P<0.001). All 6 eyes with RD involved posterior
segment IOFBs, whereas only 53% of all IOFBs were poster-
ior segment (P
= 0.028). One of the 6 eyes with preoperative
RD also had endophthalmitis. Final BCVA in this e ye was
5/200. Preoperative RD was not statistically associated with
size of entry wound or IOFB material. All 6 eyes underwent
pars plana vitrectomy, IOFB removal, endolaser, and long
4 Journal of Ophthalmology
acting tamponade (3 with C3F8 gas, 3 with silicone oil).
Three eyes also underwent encircling scleral buckles. Five
out of the 6 eyes had postoperative recurrent RDs and
required a second vitrectomy with or without sclera buckle.
Mean preoperative VA was 20/4000 (range 2/200 to HM,
mean logMAR 2.33), and postoperative last corrected VA was
20/3100 (range 10/200 to LP, mean logMAR 2.20).
Development of postoperative RD was noted in 11 of
102 eyes (11%) with no preoperative RD. Risk factors for
postoperative RD were posterior segment IOFB (P
preoperative endophthalmitis (P
= 0.001), and one or more
retinal impact sites visualized intraoperatively (P
Ten of 11 eyes involved posterior segment IOFBs, 1 of 11 eyes
involved a nail that traversed the eye and rested on the retina,
and 4 of 11 eyes had preoperative endophthalmitis. Seven of
11 eyes had IOFB impact damage on the retina with hem-
orrhage or pigmentary change visualized intraoperatively.
In comparison, only 12 of the other 46 eyes with posterior
segment IOFBs and 4 of 10 eyes with traversing IOFBs had
visible retinal impact sites. Initial surgical approach in all
11 eyes involved pars plana vitrectomy with IOFB removal.
Scleral buckling was performed in 3 eyes, and lensectomy in
of 11 eyes. One of 11 eyes required enucleation, and 2 of 11
were observed with RDs and low visual potential.
Diagnostic imaging was employed preoperatively in
77 of 108 eyes (73%). 43 of 108 eyes (41%) underwent
computerized tomography (CT). All CT scans included head
and thin (1 mm) sections through the orbits with axial and
coronal planes. Echography was performed in 45 of 108 eyes
(42%), all done by experienced echographic technicians. The
sensitivity of CT was 100% when it was obtained, detecting
12 of 12 anterior segment, 26 of 26 posterior segment,
and 5 of 5 traversing IOFBs. Echography was 98% sensitive
when it was obtained, detecting 15 of 16 anterior segment,
26 of 26 posterior segment, and 3 of 3 traversing IOFBs,
respectively. One intralenticular anterior segment IOFB was
not detected by echography. Echography reports indicated
(2%), and opacities consistent w ith endophthalmitis in 1 of
45 eyes (2%).
The presenting and postoperative visual acuities in this study
are comparable to those of prior reports [3, 6, 8, 10, 16]. Prior
studies indicate that presenting BCVA and IOFB location are
strongly correlated with ﬁnal BCVA; IOFBs that penetrate
the posterior segment have a poor visual prognosis, probably
secondary to mechanical damage to the retina and optic
nerve, as well as higher risk of endophthalmitis and RD [17,
18]. Thus, the eyes in this study were analyzed in subgroups
based on IOFB location. As previously reported, IOFBs
present anterior to or within the lens had a signiﬁcantly bet-
ter visual acuity both on presentation and after treatment
than those in the posterior segment. Intralenticular foreign
bodies can be observed for variable periods of time, as
their damage is frequently limited to cataract formation.
The “traversing” IOFBs included larger foreign objects such
as nails, wires, and ﬁsh hooks that penetrated through
the anterior segment into the posterior segment but were
partial ly protruding from the eye. This group resembled the
“posterior segment” category in involving the vitreous space
and often the retina but behaved diﬀerently in providing a
potential path of entry from the extraocular to the intraocu-
lar environment, and thus increasing the potential for endo-
phthalmitis. Otherwise, these eyes had better visual acuity
than other eyes with IOFBs restricted to the posterior seg-
ment, possibly because they involved lower velocity foreign
bodies and were frequently thin lengths of metal that could
be removed with relatively less trauma. Visual outcome over-
all was greatly aﬀected by the presence or absence of endoph-
thalmitis and RD, both of which signiﬁcantly decreased the
last corrected visual acuity. Other previously described risk
factors for poor ﬁnal visual acuity including size of IOFB and
time to IOFB removal were not found to be signiﬁcant .
Endophthalmitis has been reported in 3–11% of open-
globe injuries, and 3–17% of injuries with IOFBs [17–21].
Previous reports associate a delay in removal of the IOFB
with increased risk of endophthalmitis . Colyer et al.
reported 79 cases of endophthalmitis in eyes with combat-
related IOFBs, all of which received rapid wound closure
and systemic antibiotics but delayed removal of the IOFB,
suggesting that at least in the setting of combat, timely antibi-
otic administration may be more important than immediate
IOFB removal . Six patients (6%) with IOFB in this
study presented with endophthalmitis. Endophthalmitis eyes
trended towards a longer time interval between injury and
presentation (mean 2.9 versus 1.3 d ays), but the diﬀerence
was not statistically signiﬁcant. Posterior segment IOFB and
yard or soil-related injuries involving vegetable matter expo-
sure were risk factors for endophthalmitis, both of which
have been previously identiﬁed . None of the patients in
this study developed endophthalmitis after presentation or
surgery. A previous study reported the rate of postoperative
endophthalmitis in open-globe injuries at around 3% .
Eyes with posterior segment and traversing IOFBs—those
at highest risk for endophthalmitis—received prophylactic
broad-spectrum intravitreal antibiotics at time of surgery
in the current study, perhaps contributing to the lack of
postoperative infection. Prophylactic intravitreal antibiotics
for open-globe injuries with or without IOFBs remain
controversial [13, 20, 25, 26].
globe injuries and 6–36% of those with posterior segment
IOFBs [6, 12, 13, 27]. IOFBs are frequently associated with
hyphema, cataract, vitreous hemorrhage, and other media
opacities that limit retinal exam and make it diﬃcult to iden-
tify RDs preoperatively . The current study distinguished
preoperative RDs which were visible on exam, echography,
or in the operating room upon start of surgery from those
that developed after primary repair of the open-globe injury.
Preoperative RDs occurred in 6 eyes (6%) and were asso-
ciated with posterior segment IOFBs, endophthalmitis, and
posterior entry wounds. Postoperative RDs were likewise
associated with posterior segment IOFBs and endophthalmi-
tis, but additionally at risk in eyes with IOFB impact sites on
the retina. These areas may predispose to retinal breaks and
Journal of Ophthalmology 5
proliferative vitreoretinopathy that manifest after vitrectomy.
Wound length and IOFB size have been previously described
as risk factors for RD but were not correlated in this study
CT has become the imaging modality of choice in many
areas for screening of open-globe injuries and detection of
IOFBs. Most practice locations have access to nearby CT
machines, the scan is reasonably quick and user-independ-
ent, and newer orbital imaging protocols involve 1-2 mm sec-
tions that accurately detect open-globe injuries and IOFBs.
Echography is less commonly used in the setting of open-
globe injuries because it is a contact scan with risk of external
pressure that could gape wounds or extrude intraocular con-
tents, it requires more user experience to perform, and it does
not eﬀectively image adjacent areas like the posterior orbit,
sinuses, and intracranial space . Echography is poten-
tially advantageous for its immediate availability in many
ophthalmic clinics and its superior imaging of intraocular
pathology such as vitreous hemorrhages, RDs, and endoph-
thalmitis . In the current study, 43% of eyes underwent
CT, and 45% underwent echography. This is probably not
representative of a typical practice setting, as this study was
performed in an institution with an experienced echographic
department. Both modalities were highly s ensitive when
utilized, with CT detecting 100% of IOFBs and echography
detecting 98% in those patients. Thirty seven percent of
IOFBs in the current study were not detected on initial
clinical exam. This together with the high sensitivity of these
two imaging techniques emphasizes the need for clinical sus-
picion and appropriate ancillary testing in cases at risk for
The current study is limited by its retrospective method-
ology, the variability in the included pathology, and the
limited followup in many patients. The number of endoph-
thalmitis cases was relatively small. Multiple surgeons were
involved, and there was a spectrum of operative approaches,
particularly regarding removal of the IOFB, endotamponade
selection, and lens management.
In summary, injury in the setting of soil or vegetable mat-
ter exposure, nonmetallic IOFBs, and longer time to presen-
tation are risk factors for endophthalmitis in patients with
IOFBs. Prophylactic intravitreal antibiotics at time of IOFB
removal may lower this risk. RD is associated with posterior
segment IOFBs, posterior entry wounds, endophthalmitis,
and retinal impact sites.
This study was supported by the National Institute of Health
Center Grant no. P30-EY014801, an unrestricted grant to
the University of Miami from Research to Prevent Blindness,
New York, NY.
H. W. Flynn Jr., MD, worked as a Consultant for Alcon, Aller-
gan, Pﬁzer, Santen, and Alimara. T. Albini, MD, worked as a
Consultant for Alcon, Allergan, and Bausch and Lomb, and
they got a research support from Genentech. S. G. Schwartz,
MD, MBA, worked as a Consultant for Alimera, Bausch and
Lomb. Intellectual Property was licensed to IC Labs.
 J. P. Shock and D. Adams, “Long-term visual acuity results
after penetrating and perforating ocular injuries,” American
Journal of Ophthalmology, vol. 100, no. 5, pp. 714–718, 1985.
P. E. Liggett, “Infectious endophthalmitis after penetrating
injuries with retained intraocular foreign bodies,” Ophthal-
mology, vol. 100, no. 10, pp. 1468–1474, 1993.
management of intraocular foreign bodies,” Current Opinion
in Ophthalmology, vol. 19, no. 3, pp. 225–233, 2008.
 C. M. Greven, N. E. Engelbrecht, M. M. Slusher, and S. S. Nagy,
“Intraocular foreign bodies: management, prognostic factors,
and visual outcomes,” Ophthalmology, vol. 107, no. 3, pp. 608–
 J. J. Miller, I. U. Scott, H. W. Flynn et al., “Endophthalmitis
caused by bacillus species,” American Journal of Ophthalmol-
ogy, vol. 145, no. 5, pp. 883–888, 2008.
 L. Wickham, W. Xing, C. Bunce, and P. Sullivan, “Outcomes
of surgery for posterior segment intraocular foreign bodies—a
retrospective review of 17 years of clinical experience,” Graefe’s
Archive for Clinical and Experimental Ophthalmology, vol. 244,
no. 12, pp. 1620–1626, 2006.
 J. C. Aﬀeldt,H.W.Flynn,andR.K.Forster,“Microbialendo-
phthalmitis resulting from ocular trauma,” Ophthalmology,
vol. 94, no. 4, pp. 407–413, 1987.
 M. H. Colyer, E. D. Weber, E. D. Weichel et al., “Delayed
intraocular foreign body removal without endophthalmitis
during operations iraqi freedom and enduring freedom,” Oph-
thalmology, vol. 114, no. 8, pp. 1439–1447, 2007.
Lee, “Treatment and outcome of traumatic endophthalmitis
in open globe injury with retained intraocular foreign body,”
Ophthalmologica, vol. 224, no. 2, pp. 79–85, 2010.
 F. Kuhn and R. Morris, “Posterior segment intraocular foreign
bodies: management in the vitrectomy era,” Ophthalmology,
vol. 107, no. 5, pp. 821–822, 2000.
 A. M. Al-Omr an, E. B. Abboud, and A. M. Abu El-Asrar, “Mi-
crobiologic spectrum and visual outcome of posttraumatic
endophthalmitis,” Retina, vol. 27, no. 2, pp. 236–242, 2007.
 M. Soheilian, M. Feghi, S. Yazdani et al., “Surgical manage-
ment of non-metallic and non-mag netic metallic int raocular
foreign bodies,” Ophthalmic Surgery Lasers and Imaging, vol.
36, no. 3, pp. 189–196, 2005.
 C. Chiquet, J. C. Zech, P. Gain, P. Adeleine, and C. Trepsat,
“Visual outcome and prognostic factors after magnetic extrac-
tion of posterior segment foreign bodies in 40 cases,” British
Journal of Ophthalmology, vol. 82, no. 7, pp. 801–806, 1998.
 D. J. Pieramici, P. Sternberg Jr., T. M. Aaberg et al., “A system
for classifying mechanical injuries of the eye (globe). The
Ocular Trauma Classiﬁcation Group,” American Journal of
Ophthalmology, vol. 123, no. 6, pp. 820–831, 1997.
 D. J. Jacobs, A. Pathengay, H. W. Flynn et al., “Intravitreal
dexamethasone in the management of delayed-onset bleb-
associated endophthalmitis,” International Journal of Inﬂam-
mation, vol. 2012, Article ID 503912, 5 pages, 2012.
 M. G. L. Woodcock, R. A. H. Scott, J. Huntbach, and G. R.
Kirkby, “Mass and shape as factors in intraocular foreign body
injuries,” Ophthalmology, vol. 113, no. 12, pp. 2262–2269,
6 Journal of Ophthalmology
 H. C. Boldt, J. S. Pulido, C. F. Blodi, J. C. Folk, and T. A.
Weingeist, “Rural endophthalmitis,” Ophthalmology, vol. 96,
no. 12, pp. 1722–1726, 1989.
 Y. Zhang, M. Zhang, C. Jiang, and H. Y. Qiu, “Intraocular
foreign bodies in china: clinical characteristics, prognostic fac-
tors, and visual outcomes in 1421 eyes,” American Journal of
Ophthalmology, vol. 152, no. 1, pp. 66–73, 2011.
 N. Bhagat, S. Nagori, and M. Zarbin, “Post-traumatic infec-
tious endophthalmitis,” Survey of Ophthalmology, vol. 56, no.
3, pp. 214–251, 2011.
 R. W. Essex, Q. Yi, P. G. P. Charles, and P. J. Allen, “Post-trau-
matic endophthalmitis,” Ophthalmology, vol. 111, no. 11, pp.
tained intraocular foreign bodies and endophthalmitis,” Oph-
thalmology, vol. 97, no. 11, pp. 1532–1538, 1990.
 I. A. Chaudhry, F. A. Shamsi, E. Al-Harthi, A. Al-Theeb, E.
Elzaridi, and F. C. Riley, “Incidence and visual outcome of
endophthalmitis associated with intraocular foreign bodies,”
Graefe’s Archive for Clinical and Experimental Ophthalmology,
vol. 246, no. 2, pp. 181–186, 2008.
phthalmitis caused by moraxella species,” American Journal of
Ophthalmology, vol. 132, no. 5, pp. 788–790, 2001.
 A. Nashed, P. Saikia, W. A. Herrmann, V. P. Gabel, H. Helbig,
and J. Hillenkamp, “The outcome of early surgical repair with
vitrectomy and silicone oil in open-globe injuries with retinal
detachment,” American Journal of Ophthalmology, vol. 151, no.
3, pp. 522–528, 2011.
S. Das, “Role of prophylactic intravitreal antibiotics in open
globe injuries,” Indian Journal of Ophthalmology, vol. 51, no.
1, pp. 39–44, 2003.
 W. S. Thompson, P. E. Rubsamen, H. W. Flynn, J. Schiﬀ-
man, and S. W. Cousins, “Endophthalmitis after penetrating
trauma: risk factors and visual acuity outcomes,” Ophthalmol-
ogy, vol. 102, no. 11, pp. 1696–1701, 1995.
 T. W. Stone, N. Siddiqui, J. G. Arroyo, B. W. McCuen, and E. A.
Postel, “Primary scleral buckling in open-globe injury involv-
ing the posterior segment,” Ophthalmology, vol. 107, no. 10,
pp. 1923–1926, 2000.
 M. M. J. McNicholas, D. P. Brophy, W. J. Power, and J. F.
Griﬃn, “Ocular trauma: evaluation with us,” Radiology, vol.
195, no. 2, pp. 423–427, 1995.
 I. U. Scott, W. E. Smiddy, W. J. Feuer, and F. J. Ehlies, “The
impact of echography on evaluation and management of pos-
terior segment disorders,” American Journal of Ophthalmology,
vol. 137, no. 1, pp. 24–29, 2004.