ORIGINAL ARTICLE

Elevated white blood cell count may predict risk of orbital

implant exposure
Youn-Shen Bee, MD, PhD,*
,
Muh-Chiou Lin, MD,
*
Shwu-Jiuan Sheu, MD,*
,
John D. Ng, MD, MS, FACS

ABSTRACT RSUM
Objective: To determine the risk factors for primary implant exposure after enucleation and evisceration in infected eyes.
Design: Retrospective, comparative case series.
Participants: Patients who underwent enucleation or evisceration for infected eyes.
Methods: Records of patients who underwent enucleation or evisceration for infected eyes with placement of solid sphere implants
were reviewed. Preoperative white blood cell (WBC) count, microbiologic laboratory results, clinical features, medical treatment,
and surgical methods were recorded to evaluate the risk for implant exposure.
Results: Eighty-five infected eyes were collected. The mean age was 70.1 years. The positive culture rate was 69.4%. In 42 patients
with endophthalmitis or panophthalmitis, the most common microorganisms were Pseudomonas aeruginosa in 6 cases (20.7%)
and Klebsiella pneumoniae in 7 cases (24.1%). In 50 patients with keratitis or scleritis, the most common microorganisms were P.
aeruginosa in 14 eyes (46.7%) and Fusarium in 4 eyes (13.3%). There was a 12.9% exposure rate for the 85 patients.
Preoperative WBC count was significantly higher in patients with implant exposure compared with those without exposure (p
0.04). Preoperative WBC count more than 9500 cells/L had significantly higher exposure (p 0.001).
Conclusions: Preoperative elevated WBC count was associated with higher risk for implant exposure. Primary implantation after
enucleation or evisceration may be less safe in infected eyes with high preoperative WBC count.
Objet : Dtermination des facteurs de risque de l'exposition primaire d'un implant aprs l'nuclation et l'viscration d'yeux
infects.
Nature : Srie de cas rtrospectifs et comparatifs.
Mthodes : Examen des dossiers de patients qui avaient subi une nuclation ou une viscration d'yeux infects avec mise en
place d'implants sphriques solides. Le nombre de globules blancs (NGB) propratoires, les rsultats microbiologiques du
laboratoire, les aspects cliniques, le traitement mdical et les mthodes chirurgicales ont t pris en note pour valuer le risque
d'exposition de l'implant.
Rsultats : Quatre-vingt-cinq yeux infects ont t colligs. La moyenne d'ge tait de 70,1 ans. Le taux positif de culture tait de
69,4 %. Chez 42 patients atteints d'endophtalmie ou de panophtalmie, les microorganismes les plus communs taient le
Pseudomonas aeruginosa dans 6 cas (20,7 %) et le Klebsiella pneumoniae dans 7 cas (24.1 %). Chez 50 patients atteints de
kratite et de sclrite, les microorganismes les plus communs taient le Pseudomonas aeruginosa dans 14 yeux (46,7 %) et le
Fusarium dans 4 yeux (13,3 %). Le taux d'exposition tait de 12,9 % chez les 85 patients. Le NGB propratoire tait
significativement suprieur chez les patients avec exposition de l'implant comparativement ceux sans exposition (P=0,04). Le
NGB propratoire suprieur 9 500 cellules/L entranait une exposition suprieure (P=0,001).
Conclusions : Le NGB pr-opratoire lev tait associ un plus grand risque d'exposition de l'implant. L'implantation primaire
aprs l'nuclation ou l'viscration peut tre moins scuritaire dans les yeux infects ayant un NGB pr-opratoire lev.
Delayed or secondary orbital implantation after eviscera-
tion or enucleation for endophthalmitis has been recom-
mended to minimize risk for implant exposure and
extrusion.
1
However, primary implant placement in such
cases do not universally expose or extrude, and may avoid
prolonged hospitalization, postenucleation socket syn-
drome (i.e., ptosis, deep superior sulcus, endophthalmitis),
and/or a secondary surgical procedure with its inherent
risks for orbital hemorrhage, bleeding, infection, and so
on, as well as various anaesthetic complications with
intravenous or general anaesthesia.
2
Several factors may
contribute to spontaneous implant exposure including
poor wound closure techniques, premature suture absorp-
tion, inadequate implant placement with tissues closed
under tension (cactus syndrome), hemorrhage, infection,
and mechanical or inammatory irritation.
37
Patients
who underwent enucleation or evisceration secondary to
endophthalmitis usually have a great deal of ocular
inammation before surgery.
6
We herein evaluated the
risk factors for implant exposure in cases of infected eyes
that underwent enucleation or evisceration with primary
acrylic or silicone sphere implants.
From the *Department of Ophthalmology, Kaohsiung Veterans General
Hospital; Yuh-Ing Junior College of Health Care and Management,
Kaohsiung; School of Medicine, National Yang-Ming University,
Taipei, Taiwan; and Casey Eye Institute, Oregon Health and Science
University, Portland, OR, USA.
Originally received May 14, 2013. Final revision Aug. 11, 2013.
Accepted Sep. 13, 2013
Correspondence to Youn-Shen Bee, MD, Department of Ophthalmol-
ogy, Kaohsiung Veterans General Hospital, 386 Dazhong 1st Rd,
Kaohsiung, 813, Taiwan; ysbee@vghks.gov.tw
Can J Ophthalmol 2014;49:4549
0008-4182/14/$-see front matter & 2014 Canadian Ophthalmological
Society. Published by Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jcjo.2013.08.013
CAN J OPHTHALMOLVOL. 49, NO. 1, FEBRUARY 2014 45
METHODS
We reviewed the records of patients who underwent
enucleation or evisceration with implantation in Kaohsiung
Veterans General Hospital from January 1998 to March
2009. The medical records review was approved by the
institutional review board of Kaohsiung Veterans General
Hospital. Inclusion criteria included cases of enucleation
and evisceration with primary acrylic or silicone implanta-
tion to treat ocular infection. Cases with porous implants
were excluded to avoid the confounding factor of implant
type. Types of infection included infectious keratitis,
scleritis, endophthalmitis, and panophthalmitis. Before
enucleation or evisceration was recommended, all patients
were aggressively treated and failed to respond adequately
to topical, intracameral, intravitreous, and/or systemic
antibiotics. Some patients also underwent anterior chamber
irrigation or pars plana vitrectomy to treat the infection
before evisceration or enucleation was required. All enu-
cleations and eviscerations were performed using the same
procedural techniques for each. Evisceration patients
underwent a 360-degree conjunctival peritomy followed
by a keratectomy. The uveal contents were removed with
an evisceration spoon. Anterior sclerotomies were made
at the 10- and 4-oclock positions to allow for insertion
of an orbital implant. After implant placement, the sclera
was trimmed and closed followed by closure of Tenons
capsule and the conjunctiva. In enucleation patients, a
360-degree conjunctival peritomy was performed fol-
lowed by rectus muscle isolation suturing and trans-
ection from the globe. The superior and inferior oblique
muscles were transected and allowed to y free. After the
optic nerve was cut with scissors and the globe removed,
the orbital implant was placed posteriorly in the muscle
cone. The superior and inferior rectus muscles were
sutured to each other, and the medial and lateral rectus
muscles were sutured to the medial and lateral Tenons
capsule and conjunctiva, respectively. The anterior
Tenons capsule and conjunctiva were closed separately
with interrupted sutures. A conformer was placed and a
pressure patch was applied. Only acrylic or silicone
spheres were implanted.
We evaluated the age, sex, infection type, medical or
surgical treatment of the infection, surgical procedure used
(evisceration or enucleation), postoperative medical man-
agement, preoperative white blood cell (WBC) count and
differential count, preoperative fasting blood sugar, and
culture results as related to the implant exposure.
Statistical analysis was performed with SPSS for Win-
dows program, version 18.0 (SPSS, Chicago, Ill.). Basic
descriptive statistics were calculated on data gathered and
are reported as mean SD or n (%) as appropriate.
Differences between continuous outcome variables were
established, and putative risk factors were sought using the
MannWhitney U test and Student t test as appropriate.
Categorical data were examined using Pearsons
2
and
Fishers exact tests. All tests were 2-tailed, and p r 0.05
was considered statistically signicant.
RESULTS
We identied 262 patients who underwent enucleation
or evisceration with primary acrylic or silicone sphere
implant placement. One hundred and ninety-six patients
underwent enucleation, and 66 patients underwent evis-
ceration. Of the enucleation cases, 50 (20.5 %) were
indicated for infection, of which 30 had endophthalmitis/
panophthalmitis and 20 infectious keratitis/scleritis. Of
the evisceration cases, 35(53%) were indicated for infec-
tion, of which 12 had endophthalmitis/panophthalmitis
and 23 infectious keratitis/scleritis (Table 1).
Eighty-ve eyes in 85 patients were included in the
study; 42 patients had endophthalmitis or panophthalmi-
tis, and 43 patients had keratitis or scleritis (Table 2). The
mean age was 70.1 1.3 years. There were 54 males and
31 females. There were 37 patients with right eye involve-
ment and 48 with left eye involvement. Nine patients were
treated with topical or subconjunctival antibiotics. Fifty-
one patients received systemic antibiotics. In addition to
the medical treatment, 25 patients underwent other
invasive treatments including intravitreal antibiotic injec-
tion, debridement, pars plana vitrectomy, and amniotic
membrane transplantation. After enucleation or eviscera-
tion, 6 patients were treated with topical antibiotics and
79 patients received topical and systemic antibiotics.
Positive culture results were found in 59 patients
(64.4%), and Table 3 summarizes the culture results.
Pseudomonas aeruginosa was cultured in 20 patients
(33.9%), Klebsiella pneumoniae in 8 patients (13.6%),
Fusarium in 4 patients (6.8%), and Streptococcus pneumo-
niae in 4 patients (6.8%). Among patients with endoph-
thalmitis and panophthalmitis, 29 eyes (69%) were culture
positive. Of these the most common microorganisms were
K. pneumoniae in 7 cases (21.4%) and P. aeruginosa in
6 cases (20.7%). Among patients with infectious keratitis
Table 1Cause of infectious eye for enucleation or evisceration with primary implantation
Enucleation (N 196) Evisceration (N 66)
Infected eyes 50 (20.5%) 35 (53%)
Causes
Endophthalmitis or
panophthalmitis
Keratitis or scleritis
Endophthalmitis or
panophthalmitis
Keratitis or scleritis
Patients, n (%) 30 (15.3%) 20 (10.2%) 12 (18.2%) 23 (34.8%)
Elevated WBC and implant exposureBee et al.
46 CAN J OPHTHALMOLVOL. 49, NO. 1, FEBRUARY 2014
and scleritis, 30 eyes (69.8%) were culture positive. Of
these the most common microorganisms were P. aerugi-
nosa in 14 eyes (46.7%) and Fusarium in 4 eyes (13.3%).
Eleven patients had postoperative implant exposure, 10 of
which were culture positive. Of the patients with endoph-
thalmitis/panophthalmitis, 4 cases had postoperative
implant exposure and 2 eyes were culture positive for K.
pneumonia, 1 eye was culture positive for Proteus mirabilis,
and 1 eye was culture positive for mixed ora. Of the
patients with keratitis/scleritis, 7 had postoperative
implant exposure and 6 eyes were culture positive, 2 of
which were P. aeruginosa positive, 1 each was positive for
Fusarium, S. pneumoniae, Aspergillus, and Coxsackie A24.
Onset of implant exposure after surgery ranged from
5 to 60 days (mean 30.3 20.6 days). The preoperative
WBC count 11 108 4332 in the exposure group
patients were statistically signicantly elevated compared
with WBC count 8509 3775 in the nonexposure group
(p 0.04). Patients with preoperative WBC count more
than 9500 cells/L have statistically signicantly higher
exposure rate (p 0.001). The exposure group patients
showed higher positive culture rates (90.9%; p 0.09),
but this was not statistically signicant. All the implant
exposures occurred on the left side. There were no
signicant differences between the exposure and nonex-
posure groups in comparing the potential variable factors
of age, sex, infection site, infection type, preoperative
medical and surgical management, surgical procedure
used, neutrophil/lymphocyte cell count, and preoperative
fasting blood sugar (see Table 4).
DISCUSSION
In this study, we compared the preoperative WBC
count between the nonexposure group and exposure
group, and found a statistically signicantly higher WBC
count in the implant exposure group. However, there was
no statistically signicant difference in preoperative neu-
trophil/lymphocyte differential count. The culture-positive
rate was also higher in the exposure group, although it was
not statistically signicant. The elevated WBC counts
likely reect the severity of the infection and associated
inammation. In such cases, the risk for poor wound
healing should be highly suspected.
In our study, all cases of implant exposure involved the
left orbit. According to previous literature reviewed, we
could not nd any reports about the left eye being more
predisposed to increased severity of infection or compli-
cations compared with the right eye. The only difference
Table 2Demographic data
Endophthalmitis or Panophthalmitis Keratitis or Scleritis Total Eyes
Patients, n 42 43 85
General data
Mean age SD, y 68.7 1.9 71.4 1.9 70.1 1.3
Male/Female, n 29/13 25/18 54/31
OD/OS 18/24 19/24 37/48
Treatment modality, n (%)
Topical antibiotics 2 (4.8%) 7 (16.3%) 9 (10.6%)
Topical and systemic antibiotics 26 (61.9%) 25 (58.1%) 51 (60%)
Additional invasive procedures
*
14 (33.3%) 11 (25.6%) 25 (29.4%)
Operation, n (%)
Enucleation 30 (71.4%) 20 (46.5%) 50 (58.8%)
Evisceration 12 (28.6%) 23 (53.5%) 35 (41.2%)
Postoperative treatment
Local regimen 4 (9.5%) 2 (4.7%) 6 (6.4%)
Local and systemic regimen 38 (90.5%) 41 (95.3%) 79 (93.6%)
Exposure 4 (9.5%) 7 (16.3%) 11 (12.9%)
Exposure duration SD, d 27.5 12.1 31.9 25.0 30.3 20.6
n
Intravitreal antibiotic injection, pars plana vitrectomy, debridement, and amniotic membrane transplantation.
Table 3Cultured pathogens of infected eyes
Endophthalmitis or
Panophthalmitis/
Exposure
Keratitis or
Scleritis/
Exposure
Total/
Exposure
Patient no. 42/4 43/7 85/11
Culture positive
(rate)
29(69%)/4(100%)
30(69.8%)/6
(85.7%)
59(69.4%)/
10(90.9%)
Pathogens
Pseudomonas
aeruginosa
6 14/2 20/2
Klebsiella
pneumoniae
7/2 1 8/2
Fusarium 0 4/1 4/1
Streptococcus
pneumoniae
2 2/1 4/1
Aspergillus 1 2/1 3/1
Proteus
mirabilis
2/1 0 2/1
Mycobacterium
other than
tuberculosis
1 1 2
Streptococcus
epidermidis
1 0 1
Staphylococcus 0 1 1
Fungus 0 1 1
Coxsackie A24 0 1/1 1/1
Cladosporium 0 1 1
Aeromonas
hydrophila
1 0 1
Escherichia coli 0 1 1
Burkholderia
cepacia
1 0 1
Group B
streptococcus
1 0 1
Others 5/1 1 6/1
Elevated WBC and implant exposureBee et al.
CAN J OPHTHALMOLVOL. 49, NO. 1, FEBRUARY 2014 47
between the left and right eye is the common carotid
arteries arising from different parts of the aorta. The left
artery ows directly from the aorta, and the right common
carotid artery branches off the innominate artery. Hypo-
thetically, direct ow could provide an increased inocu-
lation of organisms to the eye, but this would apply only
for endogenous endophthalmitis cases. We have no
adequate explanation for this predisposition of left side
having higher complications than the right in our study.
In our clinic, which is a major centre, it appeared that
patients ignored their infected nondominant eyes such
that more severe infections set in before seeking medical
care. More of the population in Taiwan is right eye
dominant.
8
Therefore, it is theoretically possible that
sockets with infected left eyes were more inamed because
of delay in seeking care, thus increasing the risk for
postoperative implant exposure. However, we could not
suggest this in our study because we did not record which
eyes were dominant.
In a review of 1375 enucleations in Beijing, trauma was
the most common clinical diagnosis (62.5%), and infec-
tious or inammatory disease accounted for 1.7%.
9
In our
series, infection was a more common indication for
enucleation (20.5%) and the most common indication
for evisceration (53%). Indicators for evisceration or
enucleation vary from region to region. Tsai and Tseng
10
reported that in endophthalmitis cases in Taiwan, older
age (mean 65.6 years) was a statistically signicant risk
factor associated with the need for evisceration or enu-
cleation. Our institution is a tertiary referral centre that
draws from a 100-km radius with a large agricultural
population. The warm, moist tropical climate facilitates
the growth of fungus and other micropathogens.
11
In our
study, the mean age was 70.1 1.3 years, which would
correspond with previous reports of infectious cause
accounting for more cases for enucleation and evisceration
than reported in other studies with a more diverse age
range of reported patients.
P. aeruginosa, K. pneumoniae, and Fusarium were the
most common pathogens resulting in the need for enuclea-
tion or evisceration in our study. Endophthalmitis caused
by P. aeruginosa is associated with poor visual outcome.
12
In the Wills Eye Hospital study of eyes lost because of
microbial keratitis, Pseudomonas and Streptococcus were
cultured in 58% of cases, and there were no cases of fungal
infection in that study.
13
Margo
14
reported the relatively
high proportion of fungal infections in the hot and humid
environment of Florida. Hypothetically, the local climate
might inuence the type of pathogenic organisms causing
the predominant infectious eye diseases in a particular
region because of geographical differences in pathogen
population. In our endophthalmitis and panophthalmitis
cases who underwent enucleation or evisceration,
K. pneumoniae accounted for 24.1%, as P. aeruginosa
accounted for 20.7% of cases. In the infectious keratitis
or scleritis cases, P. aeruginosa accounted for 46.7% of cases
and Fusarium 13.3% of cases. In all infectious cases who
underwent enucleation or evisceration, P. aeruginosa was
present in 33.9% of cases, K. pneumoniae in 13.6%, and
Fusarium in 8.2%. In Southern California, Dresner and
Karesh
2
had reported that 11 patients with endophthalmitis
underwent evisceration, and 1 patient with P. aeruginosa
endophthalmitis had an implant exposure. Among the 11
extrusion cases in our series, there were 2 cases of
P. aeruginosa that had implant exposure and 2 cases of K.
pneumoniae keratitis that had implant exposure. Because of
the limited number of cases, we could not statistically
signicantly conclude whether P. aeruginosa or K. pneumo-
niae was associated with a higher exposure rate, but implant
exposure patients revealed higher positive culture rate,
although without signicant statistical difference.
Historically, many ophthalmologists believed that eviscera-
tion was the procedure of choice to treat endophthalmitis.
However, both enucleation and evisceration have been
advocated as effective treatments for such cases.
15
There is
no consensus in the current literature.
16,17
In comparing
evisceration and enucleation in terms of postoperative
exposure rate, there was no signicant difference in our study.
Shah et al.
15
reported a national survey in the United
States: Acrylic (36% in evisceration cases, 27% in enu-
cleation cases) and silicone (28% in evisceration cases,
18% in enucleation cases) appeared to be favoured in
infectious cases. Medpor (36% in evisceration cases, 55%
in enucleation cases) appeared to be favoured in cases of
eye removal for blind painful eyes. In this study, we
examined only cases of nonporous acrylic and silicone
implants with 12.9% exposure rate, and did not compare
the exposure rate between the nonporous and porous
sphere. In all our patients, the surgery method using either
evisceration or enucleation did not seem to be the risk
factor of postoperative implant extrusion.
Table 4Risk factors of implant extrusion
Nonexposure
Group
Exposure
Group p
Patients, n 74 11
Mean age SD, y 70.1 12.2 70.2 13.4 0.98
Male/Female, n 51/32 0/3 0.50
OD/OS 36/37 0/11 0.002
*
Cause

37/37 5/6 0.78

Culture positive/negative 49/25 10/1 0.09
Management (1/2/3)

7/42/18 0/7/4 0.32

Surgery enucleation/
evisceration
46/28 4/7 0.11
Hospitalization duration (Days) 16.5 13.4 19.2 12.4 0.53
WBC before surgery 8509 3775 11 108 4332 0.04
*
WBC 49000 26 7 0.062
WBC 49500 20 7 0.001
*
Differential count
Neutrophil 72.5 9.7 72.8 7.7 0.92
Lymphocyte 18.4 8.8 18.9 7.3 0.89
Fasting blood sugar 125.9 47.3 125. 9 47.4 0.99
n
p o 0.05 was considered statistically significant.

Endophthalmitis or panophthalmitis/keratitis or scleritis.

1 topical antibiotics, 2 topical and systemic antibiotics, 3 antibiotics and invasive

treatment.
Elevated WBC and implant exposureBee et al.
48 CAN J OPHTHALMOLVOL. 49, NO. 1, FEBRUARY 2014
Although enucleation or evisceration with primary
silicone or acrylic sphere implant could be considered
relatively safe for treatment of infected eyes, abnormally
elevated preoperative WBC count more than 9500 cells/L
may suggest a higher risk for postoperative implant
exposure in such cases, and delayed, secondary implant
placement may need to be considered.
Disclosure: The authors have no proprietary or commercial
interest in any materials discussed in this article.
Acknowledgements: The authors acknowledge the contribution
of Feng-Ming Hsu, MD, and Chen Wang, MD, for their
assistance in data collection.
REFERENCES
1. Shore JW, Dieckert JP, Levine MR. Delayed primary wound
closure. Use to prevent implant extrusion following evisceration
for endophthalmitis. Arch Ophthalmol. 1988;106:1303-8.
2. Dresner SC, Karesh JW. Primary implant placement with eviscera-
tion in patients with endophthalmitis. Ophthalmology. 2000;107:
1661-4.
3. Sagoo MS, Rose GE. Mechanisms and treatment of extruding
intraconal implants: socket aging and tissue restitution (the Cactus
Syndrome). Arch Ophthalmol. 2007;125:1616-20.
4. Jordan DR, Gilberg S, Mawn LA. The bioceramic orbital implant:
experience with 107 implants. Ophthal Plast Reconstr Surg.
2003;19:128-35.
5. Jordan DR, Klapper SR, Gilberg SM, et al. The bioceramic implant:
evaluation of implant exposures in 419 implants. Ophthal Plast
Reconstr Surg. 2010;26:80-2.
6. Tawk HA, Budin H. Evisceration with primary implant placement
in patients with endophthalmitis. Ophthalmology. 2007;114:1100-3.
7. Chuo JY, Dolman PJ, Ng TL, et al. Clinical and histopathologic
review of 18 explanted porous polyethylene orbital implants.
Ophthalmology. 2009;116:349-54.
8. Cheng CY, Yen MY, Lin HY, et al. Association of ocular dominance
and anisometropic myopia. Invest Ophthalmol Vis Sci. 2004;45:2856-60.
9. Cheng GY, Li B, Li LQ, et al. Review of 1375 enucleations in the
TongRen Eye Centre, Beijing. Eye (Lond). 2008;22:1404-9.
10. Tsai YY, Tseng SH. Risk factors in endophthalmitis leading to
evisceration or enucleation. Ophthalmic Surg Lasers. 2001;32:208-12.
11. Wilhelmus KR. Climatology of dematiaceous fungal keratitis. Am J
Ophthalmol. 2005;140:1156-7.
12. Eifrig CW, Scott IU, Flynn HW Jr, Miller D. Endophthalmitis
caused by Pseudomonas aeruginosa. Ophthalmology. 2003;110:
1714-7.
13. Cruz CS, Cohen EJ, Rapuano CJ, Laibson PR. Microbial keratitis
resulting in loss of the eye. Ophthalmic Surg Lasers. 1998;29:803-7.
14. Margo CE. Eyes removed for primary ulcerative keratitis with
endophthalmitis: microbial and histologic ndings. Ophthalmic Surg
Lasers. 1999;30:535-9.
15. Shah RD, Singa RM, Aakalu VK, Setabutr P. Evisceration and
enucleation: a national survey of practice patterns in the United
States. Ophthalmic Surg Lasers Imaging. 2012;43:425-30.
16. Soares IP, Franca VP. Evisceration and enucleation. Semin Oph-
thalmol. 2010;25:94-7.
17. Kahana A, Dutton J. Evisceration is useful in certain situations. Arch
Ophthalmol. 2010;128:1496.
Elevated WBC and implant exposureBee et al.
CAN J OPHTHALMOLVOL. 49, NO. 1, FEBRUARY 2014 49