Cataract development is among the most common causes of reversible vision loss in dogs.1 The surgical removal of cataracts by phacoemulsification is a common and highly successful surgical procedure in this species.2,3 Reported long-term success rates for restoration of vision in dogs following cataract surgery as derived from retrospective studies4–7 range from 90% to 93%. The most common vision-threatening complications of cataract surgery include glaucoma, retinal detachment, persistent uveitis, and corneal disease.8,9
In humans, postoperative bacterial endophthalmitis is considered a rare and severe potential complication of cataract surgery that often leads to permanent vision impairment or eye loss.10 Acute-onset postoperative infectious endophthalmitis is generally defined as that which develops within 6 weeks after ocular surgery.11 Findings of several large retrospective studies12–17 involving humans suggest that the incidence of acute-onset postoperative infectious endophthalmitis following phacoemulsification ranges from 0.03% to 0.2%. Also in humans, advanced age and systemic disease, notably diabetes mellitus, are risk factors for infectious endophthalmitis following cataract surgery.18–20 In 1 study,21 diabetic human patients had an approximately 3-fold increase in the rate of postoperative endophthalmitis after cataract surgery.
To the authors' knowledge, no estimates of the incidence of infectious endophthalmitis in dogs following cataract surgery have been reported, nor have risk factors for, and clinical characteristics of, this postoperative condition. The purpose of the study reported here was to determine the incidence of acute-onset (ie, developing ≤ 6 weeks after cataract surgery) postoperative infectious and sterile endophthalmitis in dogs following elective cataract surgery and to report clinical, microbiological, and histologic features of affected dogs.
Materials and Methods
Case selection criteria
Medical records of all dogs that underwent elective phacoemulsification for the removal of cataracts as a primary surgical procedure between July 1, 1995, and June 30, 2015, with the Cornell University Ophthalmology Service were reviewed. This period was selected because it provided complete medical records that were accessible for review. To be included in the study, dogs were required to have had a minimum of 6 weeks of follow-up (or to have developed infectious or sterile endophthalmitis prior to 6 weeks following surgery) by a veterinary ophthalmologist at Cornell University or another board-certified veterinary ophthalmologist, with documentation of the examination results in the medical record. Dogs with traumatic cataracts were excluded from the study.
Medical record review
Information was extracted from the medical records regarding dog signalment, date of surgery and last recheck ophthalmic examination, affected eye or eyes, concurrent major systemic diseases, whether an intraocular lens or other intraocular implants (eg, capsular tension ring or gonioimplant) had been placed, and any additional intraocular surgeries performed concurrently with phacoemulsification. For dogs with documented endophthalmitis, surgical complications, clinical findings, diagnostic test results, clinical disease course, treatments administered, and clinical outcome were recorded when available.
Surgical preparation and procedures
All patients received a complete physical examination, ophthalmic examination, bloodwork (serum biochemical analysis and CBC), and urinalysis prior to phacoemulsification surgery, as was routine at the hospital. Dogs with an overt systemic or ocular infection (eg, urinary tract infection, bacterial blepharitis, or bacterial conjunctivitis) identified on examination or by diagnostic test results were discharged from the hospital and treated for the infection; no cataract surgery was performed until the infection had resolved.
Presurgical preparation of dogs was similar throughout the investigative period and included clipping of the hair and cilia from the eyelids and immediate periocular skin in the anesthetic induction area. A vacuum was used to remove loose hair from the periocular region after clipping. The ocular surface and conjunctival fornices were then thoroughly flushed with sterile ocular irrigation solution. The eyelids and periocular skin, starting at the eyelid margins, were scrubbed with 5% povidone-iodine ophthalmic solution–soaked sterile gauze pads. After transportation of the dog to the operating suite, final surgical preparation of the eyelids and periocular skin was performed with 10% povidone-iodine solution sterile swab sticks. The povidone-iodine solution was allowed to dry prior to surgical draping. For immediate sequential bilateral procedures, surgery was performed on the right eye first and final surgical preparation was repeated for the left eye once the first eye surgery was complete and before the left eye was surgically draped.
Numerous surgeons performed the phacoemulsification procedures during the study period, including board-certified veterinary ophthalmologists and ophthalmology residents. Surgical approaches, techniques, and instrumentation often differed among surgeons. Pre-, intra-, and postoperative medication protocols, including the specific medications used, administration routes, and treatment durations, also differed among surgeons and time periods. In general, dogs received topically applied antimicrobials immediately prior to surgery and again for ≥ 2 weeks after surgery. All dogs received antimicrobials IV before surgeries began and at intervals while anesthetized. Most dogs also received a course of an orally administered antimicrobial following surgery. Intracameral and subconjunctival antimicrobial administration was not routinely performed during or after surgery. Tissue plasminogen activator was the only intracamerally administered medication used routinely and was given to most dogs on conclusion of surgery. Dogs were typically hospitalized for approximately 24 to 48 hours after surgery, and recheck examinations were performed at least 2 to 3 times during the subsequent 6-week period.
Clinical reexamination and sample collection
Dogs received complete ophthalmic examinations at each recheck examination, which included slit-lamp biomicroscopy, indirect ophthalmoscopy, Schirmer I tear testing, fluorescein stain application, and tonometric testing.a,b Transcorneal or transpalpebral ocular ultrasonography was performed to evaluate the posterior segment of the eye, as clinically indicated. Diagnostic samples of intraocular fluids were collected from dogs suspected to have infectious endophthalmitis for cytologic evaluation and microbial culture. Vitreocentesis and aqueocentesis were performed by use of strict aseptic technique, with dogs sedated. Samples for microbial testing were placed in blood culture bottles and evaluated by aerobic bacterial, anaerobic bacterial, and fungal culture. Samples for cytologic evaluation were stained with Wright and Gram stains. Histologic evaluation of ocular tissues was performed with H&E and Gram stains.
Statistical analysis
Descriptive statistics such as mean, SD, median, and range were calculated for continuous variables (eg, dog age and duration of clinical signs), and frequencies (%) were calculated for categorical variables (eg, dog sex, breed, and systemic diseases).
Results
Dogs
Medical record review resulted in the identification of 2,630 eyes of 1,447 dogs that underwent elective phacoemulsification during the 20-year study period and that met the inclusion criteria. Mean ± SD age of included dogs was 7.8 ± 35 years. Six hundred seventy-two (46.4%) dogs were castrated males, 612 (42.3%) were spayed females, 101 (7.0%) were sexually intact males, and 62 (4.3%) were sexually intact females. The 10 most common types of dogs were mixed-breed dog (n = 312 [21.6%]), Labrador Retriever (113 [7.8%]), American Cocker Spaniel (110 [7.6%]), Poodle (107 [7.4%]), Bichon Frise (93 [6.4%]), Miniature Schnauzer (51 [3.5%]), Boston Terrier (50 [3.5%]), Siberian Husky (39 [2.7%]), Shih Tzu (36 [2.5%]), and Yorkshire Terrier (35 [2.4%]). The most commonly recorded systemic diseases were diabetes mellitus (n = 531 [36.7%]), hypothyroidism (66 [4.6%]), hyperadrenocorticism (33 [2.3%]), and hypoadrenocorticism (5 [0.3%]).
Surgical procedures
Surgery was performed bilaterally and sequentially in 1,183 (81.8%) dogs, in the right eye only in 145 (10.0%) dogs, and in the left eye only in 119 (8.2%) dogs. Intraocular lenses were placed in 1,479 (56.2%) eyes, and 1,151 (43.8%) eyes were left aphakic. Most eyes that had been left aphakic had undergone surgery during the first years of the study period, when intraocular lenses were not being offered at the study veterinary hospital or were an elective option for dog owners. Capsular tension rings were placed during surgery in 38 (1.4%) eyes and Ahmed valve gonioimplants in 3 (0.1%) eyes. Prophylactic endoscopic cyclophotocoagulation was performed concurrently with phacoemulsification in 15 (0.6%) eyes, and a uveal mass was removed by iridocyclectomy concurrently with phacoemulsification in 1 (< 0.1%) eye.
Infectious endophthalmitis
Infectious endophthalmitis developed in 4 eyes (2 right and 2 left eyes) of 4 dogs (2 mixed-breed dogs, 1 Boston Terrier, and 1 Poodle) during the study period, representing 0.15% of included eyes and 0.28% of included dogs. Median age of affected dogs was 10.5 years (range, 9 to 11 years). All 4 dogs were castrated males. One dog had diabetes mellitus, and the other 3 dogs had no recorded systemic diseases. None of these 4 dogs required presurgical treatment of an overt systemic or ocular infection before phacoemulsification was performed.
All cases of infectious endophthalmitis were unilateral and in pseudophakic eyes. All cases developed following bilateral phacoemulsification procedures, with no additional intraocular implants or procedures performed. No clinically important intraoperative surgical complications were recorded for any affected eye. Review of historical information revealed no relationships among the dogs with infectious endophthalmitis, including no chronological associations (ie, all surgeries were separated by > 6 months).
All 4 dogs had received at least 1 recheck examination after discharge from the hospital but prior to development of clinical endophthalmitis, and each had unremarkable ophthalmic examination findings at that time. Clinical signs consistent with infectious endophthalmitis were first noted a median of 18 days (range, 11 to 32 days) after surgery and on postoperative days 11, 17, 19, and 32 for individual dogs. These findings included blepharospasm, conjunctival hyperemia, episcleral injection, diffuse corneal edema, severe aqueous flare, and miosis (Figure 1). Marked and rapidly progressive anterior chamber fibrin and hypopyon, admixed with hyphema in 2 of the dogs, was also present. One dog also had a central superficial corneal ulcer, 1 had multiple paraxial anterior stromal ulcers, and 1 had an incisional anterior stromal ulcer. Median intraocular pressure on initial evaluation was 17 mm Hg (range, 15 to 28 mm Hg).
Clinical photographs of the eyes of 3 dogs with acute postoperative bacterial endophthalmitis (A [11-year-old castrated male Boston Terrier] and B [9-year-old castrated male mixed-breed dog]) or acute postoperative sterile endophthalmitis (C [12-year-old castrated male mixed-breed dog]). Clinical ocular lesions in these dogs included conjunctival hyperemia, episcleral injection, diffuse corneal edema, aqueous flare, anterior chamber fibrin, hypopyon, and miosis.
Citation: Journal of the American Veterinary Medical Association 253, 2; 10.2460/javma.253.2.201
Clinical photographs of the eyes of 3 dogs with acute postoperative bacterial endophthalmitis (A [11-year-old castrated male Boston Terrier] and B [9-year-old castrated male mixed-breed dog]) or acute postoperative sterile endophthalmitis (C [12-year-old castrated male mixed-breed dog]). Clinical ocular lesions in these dogs included conjunctival hyperemia, episcleral injection, diffuse corneal edema, aqueous flare, anterior chamber fibrin, hypopyon, and miosis.
Citation: Journal of the American Veterinary Medical Association 253, 2; 10.2460/javma.253.2.201
Clinical photographs of the eyes of 3 dogs with acute postoperative bacterial endophthalmitis (A [11-year-old castrated male Boston Terrier] and B [9-year-old castrated male mixed-breed dog]) or acute postoperative sterile endophthalmitis (C [12-year-old castrated male mixed-breed dog]). Clinical ocular lesions in these dogs included conjunctival hyperemia, episcleral injection, diffuse corneal edema, aqueous flare, anterior chamber fibrin, hypopyon, and miosis.
Citation: Journal of the American Veterinary Medical Association 253, 2; 10.2460/javma.253.2.201
The posterior segment of the affected eye was not visible in any dog at initial evaluation, and ocular ultrasonography was performed in 3 dogs. Multiple echogenic mobile foci were identified in the vitreous body of the affected eyes in all dogs during ultrasonographic examination and were considered consistent with vitritis. No retinal detachment was identified by ocular ultrasonography in any dog.
All dogs with infectious endophthalmitis were hospitalized and received intensive medical treatment that included topical and systemic antimicrobial administration. Dogs also received various topically and systemically administered anti-inflammatory medications (ie, corticosteroid drugs and NSAIDs). Three dogs received intravitreal antimicrobial (vancomycin and amikacin sulfate) injections and anterior chamber intracameral tissue plasminogen activator injections. All eyes responded poorly to medical treatment and were surgically enucleated. Median treatment duration prior to enucleation was 4 days (range, 3 to 11 days), and individual dogs had been treated for 3, 4, 4, and 11 days before enucleation.
Vitreocentesis and aqueocentesis were performed for 3 of the 4 dogs within 24 hours after detection of clinical signs consistent with infectious endophthalmitis. Diagnostic samples were collected via strict aseptic technique while dogs were sedated, and sample collection was immediately followed by intravitreal antimicrobial and intracameral tissue plasminogen activator administration. The ocular fluid samples were divided and processed for cytologic evaluation and microbial culture. The fourth dog was taken to a veterinary ophthalmologist unassociated with Cornell University and received no diagnostic testing prior to treatment and enucleation.
Cytologic evaluation of aqueous and vitreous humor samples revealed suppurative inflammation in all 3 dogs for which this testing was performed. Neutrophils comprised > 90% of the inflammatory cells in all samples, and degenerative morphological cellular changes varied from mild to moderate. In addition to neutrophils, each sample contained < 5% macrophages and lymphocytes. Intracellular and extracellular bacterial cocci were identified in samples from 1 dog, and no infectious agents were identified in the other ocular fluid samples.
Results of microbial culture indicated all 3 cases of acute-onset postoperative infectious endophthalmitis for which testing was performed were attributable to aerobic bacteria. Staphylococcus aureus and an unidentified α-hemolytic Streptococcus sp were isolated from both aqueous and vitreous humor samples of 1 dog, methicillin-resistant Staphylococcus pseudintermedius from the vitreous humor sample of 1 dog, and an unidentified Streptococcus sp from both aqueous and vitreous humor samples of the third dog. All 4 dogs with infectious endophthalmitis had received a course of an orally administered antimicrobial following surgery (2 received amoxicillin, and 2 received cephalexin). Antimicrobial susceptibility determinations for the Staphylococcus pseudintermedius isolate indicated it was resistant to all topically and systemically administered antimicrobials given during the peri- and postoperative periods. Antimicrobial susceptibility testing revealed that the Staphylococcus aureus, α-hemolytic Streptococcus sp, and unidentified Streptococcus isolates were resistant to all topically administered antimicrobials used during the peri- and postoperative periods but susceptible to the systemically administered cefazolin used in the perioperative period and the systemically administered cephalexin given after surgery.
Histologic examination of the 4 enucleated globes from dogs with infectious endophthalmitis revealed septic, suppurative endophthalmitis in all 4 eyes (Figure 2). Suppurative inflammatory cell infiltrates were evident in the uveal tissues and anterior, posterior, and vitreous chambers. Retinal necrosis and suppurative infiltrates were identified in 3 dogs and lymphoplasmacytic retinitis and papillitis in 1 dog. In 2 eyes, focal retinal detachments with cellular infiltrates or hemorrhage in the subretinal space were identified. Nonunion of the corneal incisions associated with epithelial downgrowth and neutrophilic keratitis were identified histologically in 2 eyes and postulated to represent the route of bacterial entry into the eye. The corneal incisions were histologically unremarkable in the other 2 eyes with infectious endophthalmitis. Gram-positive cocci were identified in all enucleated globes and were situated in the vitreous body (n = 3), within the lens capsule (2), and along the corneal incision (1).
Photomicrographs of the enucleated globes of 3 dogs with acute postoperative bacterial endophthalmitis. A—The iris (asterisks) is displaced anteriorly, and the posterior chamber is expanded by neutrophilic infiltrate, fibrin, and hemorrhage. B—Marked numbers of neutrophils fill the vitreous body (lower portion of the image) and infiltrate the adjacent ciliary body (upper portion of the image). C—Numerous bacterial cocci line the inner aspect of the lens capsule (arrows), admixed with lens epithelial cells, and a neutrophilic infiltrate is visible on the anterior surface of the lens capsule. H&E stain; bar = 200 μm in all panels.
Citation: Journal of the American Veterinary Medical Association 253, 2; 10.2460/javma.253.2.201
Photomicrographs of the enucleated globes of 3 dogs with acute postoperative bacterial endophthalmitis. A—The iris (asterisks) is displaced anteriorly, and the posterior chamber is expanded by neutrophilic infiltrate, fibrin, and hemorrhage. B—Marked numbers of neutrophils fill the vitreous body (lower portion of the image) and infiltrate the adjacent ciliary body (upper portion of the image). C—Numerous bacterial cocci line the inner aspect of the lens capsule (arrows), admixed with lens epithelial cells, and a neutrophilic infiltrate is visible on the anterior surface of the lens capsule. H&E stain; bar = 200 μm in all panels.
Citation: Journal of the American Veterinary Medical Association 253, 2; 10.2460/javma.253.2.201
Photomicrographs of the enucleated globes of 3 dogs with acute postoperative bacterial endophthalmitis. A—The iris (asterisks) is displaced anteriorly, and the posterior chamber is expanded by neutrophilic infiltrate, fibrin, and hemorrhage. B—Marked numbers of neutrophils fill the vitreous body (lower portion of the image) and infiltrate the adjacent ciliary body (upper portion of the image). C—Numerous bacterial cocci line the inner aspect of the lens capsule (arrows), admixed with lens epithelial cells, and a neutrophilic infiltrate is visible on the anterior surface of the lens capsule. H&E stain; bar = 200 μm in all panels.
Citation: Journal of the American Veterinary Medical Association 253, 2; 10.2460/javma.253.2.201
Sterile endophthalmitis
In addition to the 4 dogs with infectious endophthalmitis confirmed by results of cytologic evaluation, microbial culture, and histologic evaluation, 3 other dogs (2 castrated males and 1 spayed female; 1 mixed-breed dog, 1 Bichon Frise, and 1 Labrador Retriever) that underwent phacoemulsification during the study period had clinical signs suggestive of infectious endophthalmitis. Given the cytologic, culture, and histologic findings, these dogs were deemed to have sterile endophthalmitis, representing 0.11% of all included eyes and 0.21% of all included dogs. Median age of the dogs with sterile endophthalmitis was 11 years (range, 10 to 12 years). One dog had diabetes mellitus, 1 dog had both diabetes mellitus and hyperadrenocorticism, and 1 dog had no recorded systemic diseases.
All cases of sterile endophthalmitis were unilateral (right eye in 2 dogs and left eye in 1 dog) and in pseudophakic eyes. In 2 dogs, sterile endophthalmitis developed following bilateral surgeries, and in 1 dog, it developed following unilateral surgery. All affected eyes had received no additional intraocular implants or procedures. No clinically important intraoperative surgical complications were noted for any of the eyes.
Clinical signs of endophthalmitis were first noticed a median of 21 days (range, 18 to 22 days) after surgery. All dogs had received at least 1 recheck examination after discharge from the hospital but before development of clinical endophthalmitis, and each had unremarkable ophthalmic examination findings at that time. Initial clinical findings were similar to those of the 4 dogs with infectious endophthalmitis (Figure 1). One dog was treated by immediate enucleation. Two dogs were hospitalized and received medical treatment that included topical and systemic antimicrobial administration. Various topically and systemically administered anti-inflammatory medications were also given. One dog also received intravitreal antimicrobial (ie, vancomycin and amikacin sulfate) injections. One medically treated dog responded poorly to treatment, and the globe was enucleated 3 days after treatment began. The other dog rapidly improved with medical treatment, and both the globe and vision were retained.
Vitreocentesis and aqueocentesis were performed for 1 dog with sterile endophthalmitis within 24 hours after identification of clinical findings of endophthalmitis. Vitreocentesis was performed for diagnostic purposes for the other 2 dogs with sterile endophthalmitis following enucleation surgery. Cytologic evaluation of ocular fluids revealed nonseptic suppurative inflammation. Nondegenerative neutrophils were the predominant type of leukocytes observed in all samples, with fewer numbers of macrophages, lymphocytes, and eosinophils (in 1 dog only). Aerobic, anaerobic, and fungal cultures yielded negative results for all samples from all these dogs. Histologic evaluation of the 2 enucleated globes revealed sterile suppurative endophthalmitis with no explanation for the inflammation.
Discussion
The pathogenesis of postoperative bacterial endophthalmitis involves initial intraocular bacterial penetration with subsequent spread to the posterior segment of the eye. Bacterial proliferation then triggers aggressive intraocular inflammatory responses.22 Host inflammatory processes (eg, breakdown of blood-ocular barriers, complement production, cytokine elaboration, and leukocyte chemotaxis) and bacterial virulence factors (eg, lipopolysaccharide and secretory toxins) contribute to the rapid and often dramatic damage to the delicate and poorly regenerative intraocular tissues.23
The incidence of infectious (bacterial) endophthalmitis in the dogs of the study reported here was similar to that reported for humans undergoing cataract surgery.12–17 The gram-positive bacteria recovered from the dogs (ie, Staphylococcus and Streptococcus spp) were also among the most common causative microorganisms recovered from humans with acute-onset postoperative endophthalmitis following cataract removal.24–26 A potential difference between canine and human cases of acute-onset endophthalmitis following cataract surgery is the protracted interval between surgery and initial diagnosis for dogs. In reports17,21,27 of human patients, endophthalmitis was often diagnosed earlier (ie, 6 to 10 days after surgery) than in the dogs of the present report (median, 18 days after surgery). This difference may reflect differences between dogs and humans in ocular physiologic characteristics or immunologic factors. Alternatively, dog owners may have been less likely to note the early clinical changes associated with endophthalmitis and therefore may have taken their dog for veterinary examination at a later stage of clinical infection than would happen with human patients.
The present study yielded no definitive clinical criteria that could be used to distinguish sterile endophthalmitis from bacterial endophthalmitis in dogs following cataract surgery. Both conditions were first noticed at approximately the same point after surgery and were associated with comparable clinical manifestations. A similar situation exists in humans, in which acute sterile postoperative inflammation, such as that associated with toxic anterior segment syndrome, can mimic bacterial endophthalmitis.28–29 Although microbial culture and cytologic and histologic evaluation of samples from the eyes failed to reveal microorganisms, it was also possible that some cases of endophthalmitis that were diagnosed as sterile in the present study were actually of bacterial origin. In a large study30 involving humans with clinical signs and symptoms of bacterial postoperative endophthalmitis, 18% of patients had no bacterial growth on microbial culture of aqueous humor and vitreous samples. Possible causes of sterile endophthalmitis that have been speculated, but not established, to affect dogs following phacoemulsification include endotoxin residues on surgical instruments, intraocular lens-polishing substances, viscoelastic residues, and contaminated intraocular irrigation fluids.31
In the present study, cases of delayed-onset (> 2 weeks after surgery) fibrinous uveitis or immediate-onset (< 24 hours after surgery) mild and transient hypopyon were excluded as endophthalmitis suspects. These clinical conditions are distinct from the endophthalmitis cases included in the present study, and neither clinical condition has an established relationship with an infectious process in dogs. The authors' previous diagnostic evaluations of dogs with either of these clinical conditions have also yielded no evidence of the conditions being infections or progressing to endophthalmitis. We have only rarely and sporadically observed mild hypopyon in the immediate postoperative period, and affected dogs have generally been diabetic or have had a history of chronic lens-induced uveitis; in such instances, the hypopyon has typically resolved with 1 to 3 days of medical treatment. Delayed-onset anterior uveitis, characterized by fibrin deposition in and around the capsular bag, is anecdotally reported as a common finding by many veterinary ophthalmologists and is of uncertain etiology.32
To the authors' knowledge, no studies have been reported previously that specifically investigated the frequency or characteristics of acute postoperative endophthalmitis in dogs. A report33 of histologic findings in eyes that had been enucleated or eviscerated because of complications following cataract surgery includes 18 cases of suppurative endophthalmitis. Six of those cases were associated with corneal perforation, including 4 eyes with surgical incision dehiscence and 2 with perforation distant to the corneal incision. Mean interval between cataract surgery and enucleation for dogs with endophthalmitis in that study33 was 0.92 months, and bacteria were identified by Gram stain in only 2 affected eyes. In a prior clinical study4 of general postoperative complications of cataract surgery in dogs, 4 of 290 (1.4%) eyes in which surgery was performed developed endophthalmitis, and this condition reportedly developed within 3 months after surgery for all affected dogs. Although that frequency of endophthalmitis is higher than the incidence identified in the present study, no additional details regarding the cases were provided for the other study.4 In other studies5,7,34 of postoperative complications in dogs following phacoemulsification, no cases of endophthalmitis were noted; however, those studies involved smaller sample sizes (ie, < 200 dogs) and briefer study periods than those in the present report.
During cataract surgery, bacteria may gain access to intraocular compartments in the intraoperative period through the incision or in the postoperative period through incision dehiscence, full-thickness corneal sutures, externalized vitreous strands, or other mechanisms.35 In 2 of the 4 dogs with confirmed bacterial endophthalmitis in the present study, partial corneal incision dehiscence was histologically identified, representing a likely route of intraocular migration for the bacteria. This incisional failure was not clinically apparent in 1 of the 2 dogs. In the other 2 dogs, the bacterial endophthalmitis may have resulted from bacteria that entered the eye at the time of surgery. Previous studies36,37 involving microbial culture of anterior chamber samples obtained from dogs on conclusion of cataract surgery suggest that intraocular contamination with microorganisms occurs commonly during such surgeries. Indeed, in 1 study36 involving phacoemulsification specifically, at least 1 microorganism was recovered from anterior chamber samples from 22.7% of eyes. Both bacteria and fungi were recovered in that study,36 and microorganism speciation revealed that many of the obtained isolates were identical to those from the dogs' extraocular microflora. In a different study37 in which multiple cataract surgery techniques were evaluated, 24% of anterior chamber samples had positive results of bacterial culture.
Endophthalmitis secondary to the hematogenous dissemination of bacteria to the eye could not be definitely excluded for the dogs with confirmed bacterial endophthalmitis in the present study, but this condition is uncommon in dogs and is a less plausible explanation.38–40 The bacterial species recovered from their intraocular fluids were all common inhabitants of the canine extraocular microflora, suggesting intraocular contamination with bacteria originating from the ocular surface as the crucial event initiating the development of endophthalmitis.41–43
Despite bacterial contamination of the anterior chamber being a fairly common event during cataract surgery in both dogs and humans, bacterial endophthalmitis is an uncommon sequela.36,37,44 The anterior chamber is able to clear small inocula of viable bacteria, and experimental research has shown that this ability is dependent on bacterial virulence and inoculum size.45 A larger inoculum of bacteria, or the presence of bacteria of greater virulence, is more likely to produce fulminant clinical endophthalmitis than the opposite.46 The ability of the anterior chamber to clear bacteria from the eye is assisted by an intact posterior lens capsule. An intact lens capsule presumably serves as an anatomic barrier to bacterial migration into the vitreous body. The risk of postoperative bacterial endophthalmitis in humans increases dramatically with posterior lens capsule tears, and the bacterial inoculum necessary to induce endophthalmitis is significantly reduced in this instance.47,48
The posterior lens capsule was not known to be damaged in any of the dogs that developed bacterial endophthalmitis in the study reported here; however, an ex vivo study49 involving phacoemulsification of canine eyes revealed that irrigation fluids may leak into the vitreous chamber even with an intact lens capsule. Leakage of fluid into the vitreous chamber in that study49 was promoted by higher phacoemulsification irrigation fluid inflow rates and volumes, and the investigators speculated this might have enabled microorganisms to travel from the anterior chamber to the vitreous chamber, thereby increasing the risk of postoperative endophthalmitis in the dogs.
Although a rare complication after phacoemulsification surgery, bacterial endophthalmitis was associated with a poor clinical outcome and all affected globes were ultimately enucleated in the study reported here. Eye loss occurred despite aggressive medical treatment, including prompt intravitreal antimicrobial administration, having been performed in 3 dogs. Pars plana vitrectomy is commonly recommended for humans with severe and advanced postoperative endophthalmitis to reduce bacterial loads and toxins, remove vitreous membranes, and improve distribution of intravitreally administered antimicrobials.30 Whether vitrectomy would have benefited the dogs of the present study remains unknown.
The low incidence of bacterial endophthalmitis in the present study precluded the identification of specific risk factors for this postoperative complication. A larger sample size would be needed to further characterize this uncommon infection in dogs and to determine the optimal preventative measures and therapeutic approaches to maximize outcomes for affected dogs. The role of bacterial antimicrobial resistance also requires additional study, given that the causative bacteria recovered from the dogs with infectious endophthalmitis had extensive resistance patterns (data not shown). By design, dogs with chronic or delayed-onset (ie, infection first noted > 6 weeks after surgery) postoperative endophthalmitis were excluded. This less common form of postoperative infection in humans is associated with unique clinical characteristics and etiologies,50 and evaluation of dogs for this form of endophthalmitis is warranted. Given that the underlying cause of the sterile postoperative endophthalmitis in the dogs of the present study was not identified, we also recommend further investigation into the cause of and risk factors for sterile postoperative endophthalmitis in dogs.
Footnotes
Tono-Pen, Reichert Technologies, Depew, NY.
TonoVet, Jorgensen Laboratories, Loveland, Colo.
References
1 Gelatt KN, Mackay EO. Prevalence of primary breed-related cataracts in the dog in North America. Vet Ophthalmol 2005;8:101–111.
2 Appel SL, Maggs DJ, Hollingsworth SR, et al. Evaluation of client perceptions concerning outcome of cataract surgery in dogs. J Am Vet Med Assoc 2006;228:870–875.
3 Wilkie DA, Colitz CM. Update on veterinary cataract surgery. Curr Opin Ophthalmol 2009;20:61–68.
4 Sigle KJ, Nasisse MP. Long-term complications after phacoemulsification for cataract removal in dogs: 172 cases (1995–2002). J Am Vet Med Assoc 2006;228:74–79.
5 Azoulay T, Dulaurent T, Isard PF, et al. Immediately sequential bilateral cataract surgery in dogs: a retrospective analysis of 128 cases (256 eyes). J Fr Ophthalmol 2013;36:645–651.
6 Johnstone N, Ward DA. The incidence of posterior capsule disruption during phacoemulsification and associated postoperative complication rates in dogs: 244 eyes (1995–2002). Vet Ophthalmol 2005;8:47–50.
7 Klein HE, Krohne SG, Moore GE, et al. Postoperative complications and visual outcomes of phacoemulsification in 103 dogs (179 eyes): 2006–2008. Vet Ophthalmol 2011;14:114–120.
8 Schmidt GM, Vainisi SJ. Retrospective study of prophylactic random transscleral retinopexy in the Bichon Frise with cataract. Vet Ophthalmol 2004;7:307–310.
9 Moeller E, Blocker T, Esson D, et al. Postoperative glaucoma in the Labrador Retriever: incidence, risk factors, and visual outcome following routine phacoemulsification. Vet Ophthalmol 2011;14:385–394.
10 Hashemian H, Mirshahi R, Khodaparast M, et al. Post-cataract surgery endophthalmitis: brief literature review. J Curr Ophthalmol 2016;28:101–105.
11 Vaziri K, Schwartz SG, Kishor K, et al. Endophthalmitis: state of the art. Clin Ophthalmol 2015;9:95–108.
12 Wykoff CC, Parrott MB, Flynn HW Jr, et al. Nosocomial acute-onset postoperative endophthalmitis at a university teaching hospital (2002–2009). Am J Ophthalmol 2010;150:392.e2–398.e2.
13 Ravindran RD, Venkatesh R, Chang DF, et al. Incidence of post-cataract endophthalmitis at Aravind Eye Hospital: outcomes of more than 42,000 consecutive cases using standardized sterilization and prophylaxis protocols. J Cataract Refract Surg 2009;35:629–636.
14 Freeman EE, Roy-Gagnon MH, Fortin E, et al. Rate of endophthalmitis after cataract surgery in Quebec, Canada, 1996–2005. Arch Ophthalmol 2010;128:230–234.
15 Jensen MK, Fiscella RG, Moshirfar M, et al. Third- and fourth-generation fluoroquinolones: retrospective comparison of endophthalmitis after cataract surgery performed over 10 years. J Cataract Refract Surg 2008;34:1460–1467.
16 Keay L, Gower EW, Cassard SD, et al. Postcataract surgery endophthalmitis in the United States: analysis of the complete 2003 to 2004 Medicare database of cataract surgeries. Ophthalmology 2012;119:914–922.
17 Friling E, Lundstrom M, Stenevi U, et al. Six-year incidence of endophthalmitis after cataract surgery: Swedish national study. J Cataract Refract Surg 2013;39:15–21.
18 Wong TY, Chee SP. The epidemiology of acute endophthalmitis after cataract surgery in an Asian population. Ophthalmology 2004;111:699–705.
19 Kattan HM, Flynn HW Jr, Pflugfelder SC, et al. Nosocomial endophthalmitis survey. Current incidence of infection after intraocular surgery. Ophthalmology 1991;98:227–238.
20 Hatch WV, Cernat G, Wong D, et al. Risk factors for acute endophthalmitis after cataract surgery: a population-based study. Ophthalmology 2009;116:425–430.
21 Jabbarvand M, Hashemian H, Khodaparast M, et al. Endophthalmitis occurring after cataract surgery: outcomes of more than 480,000 cataract surgeries, epidemiologic features, and risk factors. Ophthalmology 2016;123:295–301.
22 Vallejo-Garcia JL, Asencio-Duran M, Pastora-Salvador N, et al. Role of inflammation in endophthalmitis. Mediators Inflamm 2012;2012:196094.
23 Callegan MC, Booth MC, Jett BD, et al. Pathogenesis of gram-positive bacterial endophthalmitis. Infect Immun 1999;67:3348–3356.
24 Han DP, Wisniewski SR, Wilson LA, et al. Spectrum and susceptibilities of microbiologic isolates in the Endophthalmitis Vitrectomy Study. Am J Ophthalmol 1996;122:1–17
25 Kunimoto DY, Das T, Sharma S, et al. Microbiologic spectrum and susceptibility of isolates: part I. Postoperative endophthalmitis. Endophthalmitis Research Group. Am J Ophthalmol 1999;128:240–242.
26 Al-Mezaine HS, Kangave D, Al-Assiri A, et al. Acute-onset nosocomial endophthalmitis after cataract surgery: incidence, clinical features, causative organisms, and visual outcomes. J Cataract Refract Surg 2009;35:643–649.
27 Miller JJ, Scott IU, Flynn HW Jr, et al. Acute-onset endophthalmitis after cataract surgery (2000–2004): incidence, clinical settings, and visual acuity outcomes after treatment. Am J Ophthalmol 2005;139:983–987
28 Gopal L, Vijaya L. Toxic anterior segment syndrome. Br J Ophthalmol 2013;97:953.
29 Rishi E, Rishi P, Sengupta S, et al. Acute postoperative Bacillus cereus endophthalmitis mimicking toxic anterior segment syndrome. Ophthalmology 2013;120:181–185.
30 Endophthalmitis Vitrectomy Study Group. Results of the Endophthalmitis Vitrectomy Study. A randomized trial of immediate vitrectomy and of intravenous antibiotics for the treatment of postoperative bacterial endophthalmitis. Arch Ophthalmol 1995;113:1479–1496.
31 Myrna KE, Pot S, Bentley E, et al. Toxic anterior segment syndrome and are we missing it? Vet Ophthalmol 2009;12:138.
32 Wilkie DA, Colitz CM. Surgery of the lens. In: Gelatt N, Gilger BC, Kern TJ, eds. Veterinary ophthalmology. 5th ed. Ames, Iowa: Wiley-Blackwell Co, 2013;1234–1286.
33 Moore DL, McLellan GJ, Dubielzig RR. A study of the morphology of canine eyes enucleated or eviscerated due to complications following phacoemulsification. Vet Ophthalmol 2003;6:219–226.
34 Davidson MG, Nassise MP, Jamieson VE, et al. Phacoemulsification and intraocular lens implantation: a study of surgical results in 182 dogs. Prog Vet Comp Ophthalmol 1991;1:233–238.
35 Ariyasu RG, Nakamura T, Trousdale MD, et al. Intraoperative bacterial contamination of the aqueous humor. Ophthalmic Surg 1993;24:367–373.
36 Ledbetter EC, Millichamp NJ, Dziezyc J. Microbial contamination of the anterior chamber during cataract phacoemulsification and intraocular lens implantation in dogs. Vet Ophthalmol 2004;7:327–334.
37 Taylor MM, Kern TJ, Riis RC, et al. Intraocular bacterial contamination during cataract surgery in dogs. J Am Vet Med Assoc 1995;206:1716–1720.
38 Ledbetter EC, Landry MP, Stokol T, et al. Brucella canis endophthalmitis in 3 dogs: clinical features, diagnosis, and treatment. Vet Ophthalmol 2009;12:183–191.
39 Bryan LK, Clark SD, Diaz-Delgado J, et al. Rhodococcus equi infections in dogs. Vet Pathol 2017;54:159–163.
40 Massa KL, Gilger BC, Miller TL, et al. Causes of uveitis in dogs: 102 cases (1989–2000). Vet Ophthalmol 2002;5:93–98.
41 Murphy JM, Lavach JD, Severin GA. Survey of conjunctival flora in dogs with clinical signs of external eye disease. J Am Vet Med Assoc 1978;172:66–68.
42 Wang L, Pan Q, Zhang L, et al. Investigation of bacterial microorganisms in the conjunctival sac of clinically normal dogs and dogs with ulcerative keratitis in Beijing, China. Vet Ophthalmol 2008;11:145–149.
43 Prado MR, Rocha MF, Brito EH, et al. Survey of bacterial microorganisms in the conjunctival sac of clinically normal dogs and dogs with ulcerative keratitis in Fortaleza, Ceara, Brazil. Vet Ophthalmol 2005;8:33–37
44 Mistlberger A, Ruckhofer J, Raithel E, et al. Anterior chamber contamination during cataract surgery with intraocular lens implantation. J Cataract Refract Surg 1997;23:1064–1069.
45 Shockley RK, Jay WM, Fishman PH, et al. Effect of inoculum size on the induction of endophthalmitis in aphakic rabbit eyes. Acta Ophthalmol (Copenh) 1985;63:35–38.
46 Callegan MC, Gilmore MS, Gregory M, et al. Bacterial endophthalmitis: therapeutic challenges and host-pathogen interactions. Prog Retin Eye Res 2007;26:189–203.
47 Lundstrom M, Wejde G, Stenevi U, et al. Endophthalmitis after cataract surgery: a nationwide prospective study evaluating incidence in relation to incision type and location. Ophthalmology 2007;114:866–870.
48 Beyer TL, O'Donnell FE, Goncalves V, et al. Role of the posterior capsule in the prevention of postoperative bacterial endophthalmitis: experimental primate studies and clinical implications. Br J Ophthalmol 1985;69:841–846.
49 Kang S, Jeong M, Ahn J, et al. Evaluation of fluid leakage into the canine vitreous humor during phacoemulsification using contrast-enhanced magnetic resonance imaging. Vet Ophthalmol 2015;18:13–19.
50 Maalouf F, Abdulaal M, Hamam RN. Chronic postoperative endophthalmitis: a review of clinical characteristics, microbiology, treatment strategies, and outcomes. Int J Inflam 2012;2012:313248.
