Cataract surgery is not only the most common intraocular
surgery, but may well be the most common surgery performed
annually in the United States. It is estimated that
there are approximately 1.5 million cataract surgeries
performed in the United States annually. Modern cataract
extraction surgery is relatively safe with low incidence
of complications, considering the large overall number
of surgeries performed, although perioperative complications
may affect the visual rehabilitation of a significant
number of patients [1].
Perioperative complications encountered with cataract
extraction surgery can be divided into intraoperative
and postoperative. The postoperative group can be further
subdivided into early and late. Intraoperative complications
include posterior capsule rupture, with or without
vitreous loss, loss of all or part of the lens nucleus,
iris trauma and damage, Descemet's membrane detachment,
wound leak, and suprachoroidal hemorrhage.
Early postoperative complications include pupillary
block, hyphema, elevation of the intraocular pressure
associated with the use of viscoelastic substances,
persistent uveitis (with or without hypopyon), and
endophthalmitis. Late postoperative complications include
intraocular lens malposition, secondary glaucoma, cystoid
macular edema, retinal detachment, and pseudophakic
bullous keratopathy [2].
Postoperative endophthalmitis is a devastating complication
of cataract surgery. Modern advances in surgical technique
and infection prophylaxis may have reduced the incidence
of complications. Nevertheless, current reports indicate
an incidence of postoperative endophthalmitis ranging
from 0.072% [3] to 0.17% [4]. The pooled percentage
of eyes experiencing postoperative endophthalmitis
(weighted by the sample size and, when pertinent, by
quality score of the individual studies but not adjusted
for variation in duration of follow-up) was 0.13% in
a metaanalysis of current published literature [1].
This complication rate implies an incidence of about
1500 cases of postoperative endophthalmitis annually.
The outcome of treatment of endophthalmitis can vary
from useful vision to loss of the eye. Identification
of the pathogenic organisms and prompt institution
of the appropriate treatment does not always prevent
serious visual morbidity. Prevention of infectious
endophthalmitis remains the most effective means toward
better visual outcomes
Source and Access
of Pathogens
The major source of the infecting organisms
causing postoperative endophthalmitis is the patient's
own eyelid flora. Genetic analysis, using DNA isolation
and characterization, has shown an association between
the external bacterial flora and the bacteria isolated
from vitrectomy specimens in 82% of patients with postoperative
endophthalmitis [6]. The most common organisms isolated
in this study were Staphylococcus epidermidis, Staphylococcus
aureus, and Streptococcus pneumoniae,
The most common organisms found in the
eyelids as normal flora are Staphylococcus epidermidis
and Staphylococcus aureus [7, 9]. These same organisms
are also responsible for 80 to 90% of all bacterial
endophthalmitis [1-3, 6, 10] (Table 1). These observations
have dictated sterile surgical techniques and isolation
of the patient's ocular adnexa from the operative field
as important measures in the prevention of postoperative
endophthalmitis.
Complications during cataract extraction
and intraocular lens implantation are associated with
increased incidence of postoperative endophthalmitis.
Posterior capsule rupture either with or without subsequent
intraoperative anterior vitrectomy can increase the
incidence of postoperative infection by a 5-to IO-fold
(from I in 1000 to I in 100 to 200 patients) [4, II,
12]. When the capsular barrier is compromised antibiotic
prophylaxis may play an even more significant role
in reducing the incidence of postoperative infection.
In exogenous bacterial inoculation following cataract
extraction surgery, the epithelial break of the surgical
incisions appears to be the port of entry for microorganisms
into the anterior chamber [13, 14]. In modern small
incision, scleral tunnel cataract extraction surgery,
draping the conjunctiva over the scleral surgical wound
prevents contact between the scleral incision and the
nonsterile tear film. This precaution inhibits the
direct access to the scleral tunnel incision of microorganisms
that may contaminate the tear film postoperatively.
Scleral tunnel and corneal valve incisions
have gained wide acceptance among cataract surgeons.
In scleral tunneling incisions, a scleral lamellar
dissection is carried forward into clear cornea so
that the entrance into the anterior chamber forms a
watertight internal closure. The posterior placement,
narrow profile, and absence of sutures also reduce
postoperative astigmatism and improve both short- and
long-term keratorefractive stability [14, 15]. Besides
these advantages, there is potential for new infectious
complications associated with these new surgical techniques.
The scleral tunnel lies mostly in relatively nonvascular
sclera, creating a potential abscess cavity when inoculated
by an infectious organism [16]. Although rare in occurrence,
an infection of the scleral tunnel may be difficult
to access by topical antimicrobials and may require
surgical revision.
The internal corneal lip designed for
sutureless surgery acts as a oneway valve, preventing
egress of fluid from the anterior chamber. The roof
and the floor of the scleral pocket must be in tight
apposition to prevent ingress into the anterior chamber.
An infective inoculum can be introduced into the anterior
chamber by intraoperative or early postoperative fluids
ingress through the external opening of the scleral
tunnel as a result of external pressure applied to
the scleral lip. Eye rubbing, particularly with the
eye in extreme infraduction, can promote such a mechanism.
An example of this mechanism is the clinical observation
of "sterile" endophthalmitis, attributed to the postoperative
entry of the subconjunctival antibiotic into the anterior
chamber through the one-way valve incision [17]. This
influx port to the anterior chamber exists until the
surgical wound has reepithelialized. These observations
have alerted clinicians to the fact that prophylactic
subconjunctival antibiotics can conduit into the eye
in scleral tunnel sutureless procedures.
| Table I Most
Common Bacteria Responsible for Postoperative
Endophthalmitis |
 |
Staphylococcus epidermidis
Staphylococcus aureus
Streptococcus pneumoniae
Staphylococcus haemolyticus
Streptococcus sanguis
Enterobacter species (Gram-negative) |
 |
Current Concepts
Concerning Infection Prophylaxis
Careful inspection of the patient's eyelids
preoperatively will often reveal significant underlying
disease [18]. Significant preoperative blepharitis
and meibomitis raise the issue of preoperative treatment.
Lid hygiene in conjunction with topical antibiotic
ointments or systemic antibiotics can be very effective
in reducing the microbial counts of the eyelid flora
preopera- tively [19,20].
As mentioned previously the eyelid flora
harbors most of the organisms responsible for postoperative
infections [6]. Propping the patient's periocular skin
with 5% povidone iodine solution with special attention
given to propping of the eyelashes addresses the infectious
risks described above [21].
Bacterial colonization
of the patient's ocular adnexa can orginate from preexisting
prosthetic devices (contact lenses, prosthetic scleral
shell) and can act as a harbor for cataract wound contamination
[22]. Special caution is warranted for the intraoperative
isolation of such potential adnexal infectious sources,
as well as their perioperative disinfection with operative
scrubbing techniques and antimicrobial use.
Empiric prophylactic treatment has become
common practice with its aim to prevent severe ocular
morbitity associated with postoperative endophthalmitis.
Antibiotic treatment of a potential infectious inoculum
during cataract surgery is commonly used as means of
prophylaxis, despite treatment of preexisting eyelid
disease, sterilization of the operative field, and
draping of the eyelid cilia (Table 2). Routine aqueous
humor samples taken upon completion of uncomplicated
consecutive cataract extractions have yielded a positive
culture in up to 43% of patients [23, 24]. These observations
suggest that even in uncomplicated procedures with
routine sterility measures taken a significant number
of eyes are inoculated with external bacterial flora.
Surprisingly, the incidence of endophthalmitis is far
less common, suggesting that natural immune processes
can battle routine intra- and postoperative bacterial
inoculations. Postoperative prophylactic management
of cataract patients with antibiotics may eradicate
such inoculations.
| Table 2 Step-approach
to Infectious Prophylaxis |
 |
Treat underlying external disease
(blepharitis) preoperatively.
Careful disinfection of operative field and eyelid cilia preoperatively
(povidoneiodine solution).
Strict sterile operative technique.
Careful wound closure postoperatively.
Postoperative topical antibacterial prophylaxis.
Early suspicion and aggressive management of potential infection. |
 |
Perioperative Antibiotic
Prophylaxis
There is no agreement over the regimen
used in preoperative, intraoperative, or postoperative
antibiotic prophylaxis in cataract surgery. The older
literature documents a significant reduction in endophthalmitis
with the use of perioperative antibiotics [25, 26].
The use of the aminoglycosides tobramycin or gentamycin
as monotherapy following cataract extraction has been
a popular practice. Their adverse effects include epithelial
and retinal toxicity. Their activity is good against
Gram-negative organisms and Staphylococcus aureus.
They are poorly active against Staphylococcus epidermidis
and Streptococcus pneumoniae, which are responsible
for a large fraction of postoperative endophthalmitis
cases [6]. Therefore, the use of a single agent with
a specific antibacterial spectrum of action may not
provide adequate prophylaxis for postoperative infection.
Fluoroquinolones are broad-spectrum antibiotics and
can be used for endophthalmitis prophylaxis. They have
low toxicity along with a broad spectrum of antibacterial
activity. In oral form both ciprofloxacin and ofloxacin
have excellent intravitreal penetration [27-29]. Ofloxacin
has higher internal solubility that may provide an
added advantage to postoperative prophylaxis, by contributing
higher topically available concentrations and better
penetration into the eye.
Recent concern has been raised regarding
the occurrence of antibiotic resistance to these broad-spectrum
antimicrobials with their widespread use. It appears
that the small number of bacteria in the tear film,
as well as the very high concentration accomplished
with topical administration of these antibiotics, makes
resistance following ocular administration much less
likely compared to systemic use.
Subconjunctival antibiotics used at the
completion of cataract surgery have been a common form
of perioperative prophylaxis of infection since the
mid-1950s; from the beginning, their use has been controversial
[30]. Several randomized prospective studies have indicated
no advantage to topical treatment in the prevention
of postoperative infections following cataract extraction
[31, 32]. Potential toxicity with the use of these
agents is a major disadvantage of their use. Use of
subconjunctival gentamycin has been associated with
conjunctival hyperemia, edema, and capillary closure
[33]. Significant higher degrees of conjunctival injection
and anterior chamber activity have been noted with
subconjunctival injections [17, 34]. Retinal toxicity
following subconjunctival antibiotic injections has
emerged as a very significant concern and has been
suggested in several studies [35, 36]. The use of subconjunctival
antibiotics for the prophylaxis of postoperative infections
remains subject to the surgeon's personal preference.
With the recent transition of cataract
extraction from extracapsular to smaller incision phacoemulsification,
several cataract surgeons have begun the use of antibiotics
in the irrigation fluid used during phacoemulsification.
This technique has been suggested mainly by high-volume
surgeons on the basis of a small number of reports
suggesting their effectiveness [37, 38]. The antibiotics
used are commonly an aminoglycoside or a cephalosporine.
Continuous flow of antibiotic during phacoemulsification
intraocularly does provide the theoretical advantage
of sterilization of any infectious inoculum, although
in practice the amount of antibiotic that is routinely
used may be insufficient for this purpose [38A]. However,
the exposure of intraocular tissues and subsequent
toxicity would depend on the duration and amount of
irrigation fluid used during the procedure, parameters
that are not quite standardized for each individual
surgery and surgeon. There is potential for extensive
toxicity of the corneal endothelium, trabecular meshwork,
and retina, especially in cases where the posterior
capsule is compromised. The potential for dosage error
in the irrigation fluid preparation can result in devastating
complications. In most cases the actual preparation
of the irrigation fluid/antibiotic mixture is not even
done by the surgeon.
Intraoperative antibiotic administration
in the irrigation fluid during phacoemulsification
is another form of perioperative prophylaxis with questionable
efficacy and serious potential complications. Their
use remains again to the preference and clinical experience
of the individual surgeon, although the available literature
is not quite convincing.
Diagnosis and Management
of Postoperative Infections
The diagnosis of endophthalmitis is suspected
in any patient with excessive intraocular inflammation
following cataract extraction. The predominant symptoms
are decreased vision and pain [39]. Once the diagnosis
of endophthalmitis is suspected prompt management can
improve the prognosis [40]. Postoperative Propionibacterium
endophthalmitis is a form of delayed-onset infection
often confused with persistent postoperative inflammation
in the past. Recent reports have increased the clinical
awareness for this entity where surgical management
is associated with little recurrence potential and
relatively good prognosis [41, 42]. As determined by
the outcome of endophthalmitis management, early suspicion
and diagnosis plays an essential role in the eventual
visual rehabilitation of the patient.
Current guidelines consist of a complete
ophthalmic evaluation including ophthalmoscopy and
posterior segment ultrasonography if needed [43]. Intraocular
cultures, taken at that point may confirm the diagnosis
and guide medical treatment. Vitreous cultures obtained
by pars plana vitrectomy or vitreous aspirate appear
to have a higher positive yield than anterior chamber
cultures alone [44]. Inoculation of the aqueous or
vitreous specimen on multiple culture media may optimize
the culture yield. Customarily, the media used are
blood agar and chocolate agar for identification of
bacteria, Sabouraud's agar for isolation of fungal
organisms, and thioglycolate breath for isolation of
anaerobic organisms.
Pars plana vitrectomy can be utilized
as a diagnostic as well as a treatment remedy in the
management of endophthalmitis. The theoretical advantages
of vitrectomy in the management of infectious endophthalmitis
include the debulking of the infectious load and removal
of toxins, as well as the improved penetration and
distribution of the intraocular medications as they
can be administered intravitreally during the procedure.
Disadvantages include the risks and complications associated
with any vitrectomy [45]. The consensus regarding therapeutic
vitrectomy, prior to the Endophthalmitis Vitrectomy
Study, was utilizing pars plana vitrectomy when there
is extensive vitreous inflammation precluding the visualization
of the retina and also impairing significantly the
patient's visual acuity [46].
Although not the primary focus of this
article, the antibiotic treatment of endophthalmitis
currently consists of intravitreal antibiotic administration.
The most popular antibiotic regimen consists of ceftazidime
(2.25mg/O.1 ml) in conjunction with vancomycin (1 mg/O.1
ml) with or without the addition of intravitreal corticosteroid
administration (dexamethasone 0.4 mg/O.1 ml). It appears
that culture-negative and Staphylococcal endophthalmitis
have better prognosis than Streptococcal, Hemophilus,
or Bacillus infections [18, 43, 44, 46].
The prognosis of postoperative endophthalmitis
is closely related to the type of infective organism
as noted above, as well as the length of time between
onset of infection and initiation of therapy. Traditionally,
the visual outcomes are divided in better than 20/400
or worse than 20/400 visual acuity. Overall success
of treatment of postcataract surgery endophthalmitis
indicates that vision of equal to or better than 20/400
is achieved in 44% to 73% of eyes [40, 43].
Finally, the results of the recent Endophthalmitis
Vitrectomy Study suggest that the use of systemic antibiotics
along with intravitrea does not affect the prognosis
and could be omitted. Also, immediate vitrectomy after
the diagnosis of endophthalmitis benefited only patients
with poor visual acuity (light perception) at the time
of diagnosis [47]. In this multicenter prospective
study of 420 patients, the intravitreal antibiotics
used were amikacin (0.4 mg in O.I ml, and vancomycin
1.0 mg in O.I ml). The systemic antibiotics used in
this trial, however, are not effective against Staphylococcus
epidermidis, the most common cause of postoperative
endophthalmitis. This study suggests that the omission
of systemic antibiotic treatment and, for that reason,
inpatient care, can reduce toxic effects and the cost
of care. Also, immediate vitrectomy in patients with
visual acuity better than light perception is probably
not necessary, but may offer a substantial benefit
to the patients that present with light perception-only
vision. In summary, postoperative endophthalmitis can
be a devastating complication of cataract surgery.
Recent advances in surgical technique have introduced
new mechanisms for pathogenic organisms to cause infection.
The infectious source and its access into the eye are
now better understood. This review focused on current
concepts involving the incidence, etiology, intraoperative
and postoperative risks, antibiotic prophylaxis and
diagno- sis, and management of infection following
cataract surgery. Additional research may clarify and
improve the role of prophylactic intraoperative and
postoperative antibiotic administration.
References
- Powe NR, Schein OD, Gieser SC, et aL Synthesis
of the literature on visual acuity and complications
following cataract extraction with intraocular lens
implantation. ArchOphthalmol 1994;112:239-252
- Goodman DF, Stark WJ, Gottsch JD. Complications
of cataract extraction with intraocular lens implantation.
Ophthalmic Surg 1989;20(2): 132-140
- Kattan HM, Flynn HW, Pflugfelder SC, et al. Nosocomial
endophthalmitis survey. Ophthalmology 1991;98:227-238
- Javitt JC, Vitale S, Tanner JK, et al. National
outcomes of cataract extraction: Endophthalmitis
following cataract surgery. Arch Ophthalmol 1991;19:1085
- OlsonJC, Flynn HW, Forster RK. Results in the
treatment of postoperative endoph- thalmitis. Ophthalmology
1983;90:692-699
- Speaker MG, Milch FA, Shah MK, et al. Role of
the external bacterial flora in the pathogenesis
of acute postoperative endophthalmitis. Ophthalmology
1991;98: 639-650
- DunningtonJH, Locatcher-Khorazo D. Value of cultures
before operation for cata- ract. Arch Ophthalmol
1945;87:215-219
- Allansmith MR, Anderson RP, Butterworth M. The
meaning of preoperative cul- tures in ophthalmology.
Trans Am Acad Ophthalmol Otolaryngol 1969; 73: 683-690
- Groden LR, Murphy B, RodniteJ, et al. Lid flora
in blepharitis. Cornea 1991; 10: 50
- MenikoffJA, Speaker MG, Marmor M, et al. A case
control study of risk factors for postoperative endophthalmitis.
Ophthalmology 1991;12:1761-1768
- Kattan HM, Flynn HW, Pflugfelder SC, et al. Nosocomial
endophthalmitis survey: Current incidence of infection
after intraocular surgery. Ophthalmology 1991;98:
227
- Maxwell DP, Diamond JG, May DR. Surgical wound
defects associated with endoph- thalmitis. Ophthalmic
Surg 1994:25(3): 157-161
- Bauman WC, D'Amico DJ. Surgical techniques in
diagnosis and management of suspected endophthalmitis.
Int Ophthalmol Clin 1992;32:113-128
- Masket S. Keratorefractive aspects of the scleral
pocket incision and closure method for cataract surgery.
J Cataract Refract Surg 1989; 15:70- 77
- Armeniades CD, Boriek A, Knolle GE. Effect of
incision length, location, and shape on local corneoscleral
deformation during cataract surgery. J Cataract Refract
Surg 1990;16:83-87
- Ormerod LD, Pukiin JE, McHenry JG, et al. Scleral
flap necrosis and infectious endophthalmitis after
cataract surgery with a scleral tunnel incision.
Ophthalmology 1993:100:159-163
- Nelson DB, Donnenfeld ED, Perry HD. Sterile endophthalmitis
after sutureless cataract surgery. Ophthalmology
1992;99:1655
- McCulleyJP, DoughertyJM, Deneau DG. Classification
of chronic blepharitis. Oph- thalmology 1982;89:1173
- McCulley JP, Dougherty JM. Blepharitis associated
with acne rosacea and sebor- rheic dermatitis. Int
Ophthalmol Clin 1985;25:159
- Hyndiuk RA, Burd EM, Hartz A. Efficacy and safety
of mercuric oxide in the treatment of bacterial blepharitis.
Antimicrob Agents Chemother 1990; 34: 610
- Speaker MG, MenikoffJA. Prophylaxis of endophthalmitis
with topical povidone- iodine. Ophthalmology 1991;98:1769-1775
- Morris R, Camesasca Fl, Byrne J, et al. Postoperative
endophthalmitis resulting from the prosthesis contamination
in a monocular patient. AmJ Ophthalmol 1993; 116:346-349
- Ariyasu RG, Dugel PU, Nakamura T, et al. Postoperative
bacterial contamination of the aqueous humor. Invest
Ophthalmol Vis Sci 1992;(Suppl):36-42
- Dickey JB, Thompson KD, Jay WM. Anterior chamber
aspirate cultures after un- complicated cataract
surgery. AmJ Ophthalmol 1991; 112:278
- Meredith TM. Prevention of postoperative infection.
Arch Ophthalmol 1991:109: 944-945
- Alien HE, Mangiaracine AB. Bacterial endophthalmitis
after cataract extraction, II: Incidence in 36,000
consecutive operations with special reference to
preoperative antibiotics. Arch Ophthalmol 1974 ;91:3
- El Baba FZ, Trousdale MD, Gauderman WJ, et al.
Intravitreal penetration of oral ciprofloxacin in
humans. Ophthalmology 1992;99:483-486
- Perry HD, Donnenfeld ED, Snyder R, et al. Intracorneal,
aqueous humor and vitreous humor penetration of topical
ofloxacin. Invest Ophthalmol Vis Sci 1994(Suppl);35:1674
- Donnenfeld ED, Schrier A, Perry HD, et al. Penetration
of topically applied cipro- floxacin, norfloxacin,
and ofloxacin into the aqueous humor. Ophthalmology
1994; 101:902-905
- McMillanJJ, Mead MD. Prophylactic subconjunctival
antibiotics after cataract ex- traction-evaluation
of their desirability and efficacy. Int Ophthalmol
Clin 1994; 34(3):43-49
- Chalkley THE, Shoch D. An evaluation of prophylactic
subconjunctival antibiotic injection after cataract
surgery. AmJ Ophthalmol 1967;64:1084-1087
- Aronstam RH. Pitfalls of prophylaxis. AmJ Ophthalmol
1964:57:312-315
- Jenkins CDG, McDonnel PJ, Spalton DJ. Randomized
single blind trial to compare the toxicity of subconjunctival
gentamicin and cefuroxime in cataract surgery. Br
J Ophthalmol 1990:74:734-738
- Sanders R, Macewen CJ, Haining WM. A comparison
of prophylactic, topical, and subconjunctival treatment
in cataract surgery. Eye 1992:6:105-110
- Campochiaro PA, Conway BP. Aminoglycoside toxicity-a
survey ofretinal special- ists. Arch Ophthalmol 1991:109:946-949
- McDonald HR, Schatz H, Alien AW, et al. Retinal
toxicity secondary to intraocular gentamicin injection.
Ophthalmology 1986:93:871-877
- Gills JP. Filters and antibiotics in irrigating
solution for cataract surgery. J Cataract Refract
Surg 1991:17:385
- Schneider DM, Gills JP, Oyarzu NM. What specific
means of antibacterial prophy- laxis do you recommend
before, during, and after routine cataract surgery?
J Cata- ract Refract Surg 1993:19:108-111
38A. Gritz DC, Cevallos AV, Smolin G, et al. Antibiotic supplementation of
intraocular irrigating solutions: an in vitro model of antibacterial action.
Ophthalmology 1996:103(8): 1204-1209
- Ormerod LD, Becker LE, Cruise RJ, et al. Endophthalmitis
caused by the coagulase- negative Staphylococci.
Ophthalmology 1993; 100:724-729
- Bohigian GM, Olk RJ. Factors associated with poor
visual result in endophthalmitis. Amj Ophthalmol
1986:110:332-334
- Winward KE, Pflugfelder SC, Flynn HW, et al. Postoperative
Propionibacterium endophthalmitis. Treatment strategies
and long-term results. Ophthalmology 1993; 100:447-451
- Meisler DM, Palestine AG, Vastine DW, et al. Chronic
Propionibacterium endoph- thalmitis after extracapsular
cataract extraction and intraocular lens implantation.
ArnJ Ophthalmol 1986; 102:733-739
- Meredith TA. Vitrectomy for infectious endophthalmitis.
In: Ryan SJ, ed. Retina. St. Louis: C.V. Mosby Co,
1989:601-609
- Driebe WT, Mandelbaum S, Forster RK, et al. Pseudophakic
endophthalmitis; Diag- nosis and treatment. Ophthalmology
1986;93-.442-448
- Peyman GA, Raichand M, Bonnet TO. Management of
endophthalmitis with pars plana vitrectomy. BrJ Ophthalmol
1980:21:332-346
- Kremer PA, Abbott RL. Management of endophthalmitis.
Ophthalmol Clin North Am 1994;7(l):39-50 47. Endophthalmitis
Vitrectomy Study Group. Results of the endophthalmitis
vitrec- tomy study. Arch Ophthalmol 1995:113:1479-1496
Anastasios John Kanellopoulos, MD
Source: International Ophthalmology Clinics,
Summer 1996
|