Acute-onset Endophthalmitis After Cataract Surgery (2000 –2004): Incidence, Clinical Settings, and Visual Acuity Outcomes After Treatment JOHN J. MILLER, MD, INGRID U. SCOTT, MD, MPH, HARRY W. FLYNN, JR., MD, WILLIAM E. SMIDDY, MD, JEAN NEWTON, RN, AND DARLENE MILLER, MPH
● PURPOSE:
To report the incidence, clinical settings, and visual acuity outcomes of acute-onset endophthalmitis after cataract surgery. ● DESIGN: Retrospective, observational case series. ● METHODS: Annual cataract surgery statistics were determined by review of electronic surgical records. The clinical and microbiologic records were reviewed of all patients with clinically diagnosed endophthalmitis within 6 weeks after cataract surgery at a single universityaffiliated hospital between January 2000 and November 2004. MAIN OUTCOME MEASURES: Operative technique, intraoperative complications, and visual acuity. ● RESULTS: The incidence of acute-onset endophthalmitis after cataract surgery was 0.04% (7/15,920) for cataract surgeries of all methods, 0.05% (6/11,462) for cataract surgery by clear cornea phacoemulsification, and 0.02% (1/4,458) for cataract surgery by methods other than clear cornea phacoemulsification (P ⴝ .681, Fisher’s exact test). Six of seven (86%) cases occurred in the right eye, and all cases were performed by right-handed surgeons through temporal incisions. Five of seven (71%) patients had relative immune compromise. Four of seven (57%) patients had an intraoperative complication: vitreous loss in three patients and iris prolapse in one patient. Two patients had topical placement of lidocaine 2% gel before povidone-iodine preparation. The visual acuity at final follow up was 20/25 or better in four patients and count fingers or worse in three patients.
● CONCLUSIONS:
The incidence of acute-onset endophthalmitis after temporal clear cornea incision phacoemulsification is low (0.05%). Potential risk factors for endophthalmitis may include intraoperative complications, relative immune compromise, application of lidocaine 2% gel before povidone-iodine preparation, and inferior incision location. (Am J Ophthalmol 2005; 139:983–987. © 2005 by Elsevier Inc. All rights reserved.)
E
NDOPHTHALMITIS IS AN UNCOMMON BUT SERIOUS
complication of intraocular surgery that may result in severe vision loss. The increasing use of temporal clear cornea incisions and sutureless surgery may change the incidence of acute-onset endophthalmitis after cataract surgery. However, the incidence of endophthalmitis after cataract surgery performed at the Bascom Palmer Eye Institute has not increased from the mid-1980s to the turn of the millennium: 1984 to 1989 (0.07%),1 1990 to 1994 (0.09%),1 and 1995 to 1999 (0.06%).2 The current study investigates the incidence and clinical settings of acute-onset endophthalmitis after cataract surgery in the new millennium and assesses visual acuity outcomes after treatment among patients who developed endophthalmitis after cataract surgery at a universityaffiliated hospital.
METHODS THE STUDY PROTOCOL WAS APPROVED BY THE INSTITU-
Accepted for publication Jan 20, 2005. From the Department of Ophthalmology, Bascom Palmer Eye Institute, University of Miami School of Medicine, Miami, Florida. Supported in part by Research to Prevent Blindness, Inc., New York, New York. Inquiries to Ingrid U. Scott, MD, MPH, Bascom Palmer Eye Institute, PO Box 016880, Miami, FL 33101; fax: (305) 326-6417; e-mail:
[email protected] 0002-9394/05/$30.00 doi:10.1016/j.ajo.2005.01.025
©
2005 BY
tional Review Board of the University of Miami, School of Medicine. The study was a retrospective, observational case series. Annual cataract surgery statistics were determined by review of electronic surgical records. Surgeries were categorized as either clear cornea phacoemulsification or other methods of cataract surgery
ELSEVIER INC. ALL
RIGHTS RESERVED.
983
RESULTS
TABLE 1. Endophthalmitis After Cataract Surgery: Annual Incidence
Year
Number of Cases/Number Cataract Surgeries
Incidence (%) of Endophthalmitis
2000 2001 2002 2003 2004* Total
2/2,949 1/3,162 0/3,434 3/3,531 1/2,844 7/15,920
0.07 0.03 0.00 0.10 0.04 0.04
THE 5-YEAR INCIDENCE RATE OF ACUTE-ONSET ENDOPH-
thalmitis after cataract surgery was 0.04% (7/15,920) for cataract surgeries of all methods, 0.05% (6/11,462) for cataract surgery by a temporal clear cornea approach to phacoemulsification, and 0.02% (1/4,458) for cataract surgery through methods other than clear cornea phacoemulsification (P ⫽ .681, Fisher’s exact test). The incidence of endophthalmitis by year is displayed in Table 1. The median age was 74 years (range: 50-83 years). Six of seven (86%) endophthalmitis cases occurred in right eyes, and all cases were performed by right-handed surgeons. Dictated operative reports indicated a slightly inferior temporal clear cornea incision location in all cases, and clinical photographs, which were available in three of seven (43%) patients, confirmed the same location. At the time of cataract surgery, sutures were placed to reapproximate the incision in three patients; a single interrupted nylon suture was placed through the clear cornea phacoemulsification incision in patients 5 and 6, whereas multiple interrupted sutures closed the larger corneal wound in patient 7. No wound leaks were noted on the first postoperative day. Furthermore, the lack of hypotony on the first postoperative day (intraocular pressure ranged from 11-25 mm Hg) provides further evidence that there were no significant occult wound leaks. Clinical data are summarized in Table 2. The mean duration from day of cataract surgery to day of diagnosis with endophthalmitis was 10 days (range: 1-21 days). Five of the seven vitreous isolates produced culture positive growth: four coagulase-negative Staphylococcus and one Streptococcus pneumoniae. The S. pneumoniae case presented on the first postoperative day, whereas the coagulase-negative Staphylococcus cases presented at a median postoperative day 11 (range: 4-21 days). The two culturenegative cases presented on postoperative days 5 and 16. The number and type of intraocular lenses are as follows: four acrylic foldable posterior chamber lenses, two acrylic injectable posterior chamber lenses, and one polymethyl methacrylate anterior chamber lens. Four of the seven cases (57%) had an intraoperative complication: three with vitreous loss and one with iris prolapse. One of these complicated cases was converted from clear cornea phacoemulsification to extracapsular extraction by enlarging the temporal clear cornea incision. Of the three remaining cataract operations performed without intraoperative complication, topical 2% lidocaine gel was used in two (67%) before povidone-iodine preparation (Table 2, cases 2 and 3). One of the 3 uncomplicated cases occurred in a patient with polymyalgia rheumatica who had been taking the immunosuppressive agents mycophenolate and etanercept for several years. Four of seven (57%) patients had diabetes mellitus. Perioperative antibiotics were selected at the discretion of each surgeon and are summarized in Table 3. Although
*January through November 2004.
(phacoemulsification through scleral tunnel, extracapsular cataract extraction, or intracapsular cataract extraction). This heterogeneous group also included surgeries initiated as clear cornea phacoemulsification but later converted to extracapsular cataract extraction through either an enlarged temporal corneal incision or through a superior limbal incision. Clinical records were reviewed of all cataract surgery patients who developed acute-onset postoperative endophthalmitis, defined as clinically diagnosed endophthalmitis that occurred within 6 weeks of cataract surgery. Both culture positive and culture negative cases were included. All infections after an operative procedure are reported to the Infection Control Committee of the Bascom Palmer Eye Institute. As part of operating room protocol during the time of the study (2000-2004), povidone-iodine solution was used to prepare the lids, lashes, and conjunctiva before cataract surgery. If a patient reported a history of iodine allergy, 3% hexachlorophene was used. No antibiotics were placed in the surgical infusion fluid during the study period. The diagnosis of endophthalmitis was based on decreased visual acuity and typical clinical features, including marked intraocular inflammation. In all clinically diagnosed patients, anterior chamber and/or vitreous cultures were obtained, and intravitreal antibiotics were administered on the day of diagnosis. Stored bacterial isolates from culture-positive cases were tested in vitro for sensitivity to vancomycin, ceftazidime, gentamicin, ciprofloxacin, ofloxacin, levofloxacin, gatifloxacin, and moxifloxacin by disk diffusion method. Zones of inhibition were measured after 24, 48, and 72 hours of incubation at 35 C. Cataract surgeries that were combined with any other procedure, including penetrating keratoplasty, pars plana vitrectomy, or trabeculectomy, were excluded from the study. Patients with delayed-onset endophthalmitis (infection diagnosed later than 6 weeks after surgery) and endophthalmitis referred to the Bascom Palmer Eye Institute after cataract surgery performed elsewhere were excluded from the current study. 984
AMERICAN JOURNAL
OF
OPHTHALMOLOGY
JUNE 2005
TABLE 2. Clinical Settings, Treatment, and Visual Acuity Outcomes of Acute-onset Endophthalmitis After Cataract Surgery
Patient
Eye (Age)
Anesthesia
1
OD (83)
Retrobulbar
2
OD (50)
Topical*
3
OD (82)
Topical*
4
OD (52)
Topical
5
OD (74)
Peribulbar
6
OD (68)
Retrobulbar
7
OS (78)
Topical
Intraocular Lens
Foldable acrylic Injectable acrylic Injectable acrylic Foldable acrylic Rigid PMMA Foldable acrylic Foldable Acrylic
Complication
Days to Diagnosis
None
5
None
13
None
1
Iris prolapse
4
Vitreous loss Conversion to ECCE Vitreous loss Displaced lens fragments Vitreous loss Displaced lens fragments
9
Initial Management (Medication)
Organism Cultured
Tap & inject (V, C, Dexa) PPV (V, C, Dexa)
21
Tap & inject (V, C, Dexa) Tap & inject (V, C, Dexa) Tap & inject (V, C, Dexa) PPV (V, C, Dexa)
16
PPV (V, C, Dexa)
Final Acuity (Follow-up months)
None
LP (22)
Staphylococcus epidermidis Streptococcus pneumoniae Staphylococcus auricularis Staphylococcus epidermidis Staphylococcus epidermidis
20/20 (12)
None
20/20 (49)
HM (2) 20/20 (7) CF (19) 20/25 (14)
TABLE 3. Perioperative Antibiotics in Patients With Endophthalmitis After Cataract Surgery and Antibiotic Resistance of Individual Isolates Patient
Preoperative Topical
1
Ofloxacin QID ⫻ 2 days
2 3
Gentamicin 2 hours preoperatively None
4
5
Intraoperative (location)
Postoperative Topical
Comment*
Ceftazidime 50 mg (subconjunctival) None
Ofloxacin QID
Culture negative
Moxifloxacin QID
Polymyxin B & trimethoprim (topical)
Polymyxin B & trimethoprim, QID
None
Ofloxacin (topical)
Ofloxacin QID
Ofloxacin (topical)
Ofloxacin QID
6
Gentamicin 2 hours preoperatively None
Ofloxacin QID
7
Ofloxacin QID ⫻ 2 days
Ceftazidime 50 mg (subconjunctival) Gentamicin (topical)
Sensitive to ceftazidime, gentamicin, and fluoroquinolones Sensitive to ceftazidime and fluoroquinolones; resistant to gentamicin Sensitive to gentamicin and fluoroquinolones; resistant to ceftazidime Sensitive to gentamicin; resistant to ceftazidime and fluoroquinolones Resistant to gentamicin, ceftazidime, and fluoroquinolones Culture negative
Ofloxacin QID
All patients received povidone-iodine preparation to lids, lashes, and conjunctiva in the immediate preoperative period. No antibiotics were used in the infusion fluid during cataract surgery. *All isolates were sensitive to vancomycin.
all patients received a povidone-iodine preparation to the lids, lashes, and conjunctiva, no preoperative antibiotic was used in three patients. Of the four patients who received preoperative antibiotics, two received ofloxacin and two received gentamicin. In vitro testing demonstrated that two of the five (40%) bacterial isolates in this study were resistant to all tested fluoroquinolones, including commercially available fourth generation fluoroquinolones. Two of five (40%) were resistant to gentamicin, and three of five (60%) were resistant to ceftazidime. None of VOL. 139, NO. 6
the isolates was resistant to vancomycin. On the day of endophthalmitis diagnosis, each patient received intravitreal vancomycin 1 mg, ceftazidime 2.25 mg, and dexamethasone 0.4 mg. Of the six patients with endophthalmitis after a temporal clear cornea approach to phacoemulsification, four achieved a final visual acuity of 20/25 or better, and two had a final visual acuity of hand motions or worse. One of these patients had a poor visual outcome due to endophthalmitis-related retinal detachment, whereas the other
ENDOPHTHALMITIS FOLLOWING CATARACT SURGERY
985
TABLE 4. Reported Incidence of Endophthalmitis After Cataract Surgery in Selected Published Series
First Author
Location
Duration of Study
# Cases/Total
Incidence of Endophthalmitis
Comment
Semmens12 Kamalarajah13 Mayer9 Wong14 Nagaki5 Colleaux6 Current series
Australia UK UK Singapore Japan Canada BPEI
1980–1998 10/1999–9/2000 1991–2001 1996–2001 1998–2001 9/1994–1/1998 2000–2004
188/94,653 196/230,000 30/18,191 34/44,803 15/12,317 10/13,886 7/15,920
0.20% 0.09% 0.16% 0.08% 0.12% 0.07% 0.04%
No apparent increase with clear cornea UK National Health Service Survey Decreased incidence with injectable IOLs 8-fold higher incidence with capsule tear 4.6-fold higher incidence with clear cornea No significant increase with clear cornea No significant increase with clear cornea
BPEI ⫽ Bascom Palmer Eye Institute; IOL ⫽ intraocular lens; UK ⫽ United Kingdom.
had a poor outcome due to phthisis bulbi. The patient with endophthalmitis after a converted extracapsular cataract surgery via temporal clear cornea incision had a final visual acuity of counting fingers attributable to retinal detachment with associated proliferative vitreoretinopathy.
not described in this report. In the current series, 86% of endophthalmitis cases occurred in the right eye, where the right-handed surgeons placed the corneal incision inferotemporally. Inferiorly placed filtering blebs have been associated with an increased risk of both acute- and delayed-onset endophthalmitis, perhaps associated with the proximity of the bleb to the inferior tear lake and inferior lid margin.7 The most commonly used preoperative antibiotics in the current study were gentamicin and ofloxacin. At the present time, the fourth generation fluoroquinolones are commonly used for prophylaxis during cataract surgery and have a broad range of coverage for both grampositive and gram-negative bacteria. However, resistant organisms may be encountered, including two of five (40%) isolates in the current study. None of the five gram-positive isolates was resistant to vancomycin, confirming its continued role as initial endophthalmitis treatment. A higher incidence of endophthalmitis has been associated with foldable (1.21%) vs injectable (0.028%) intraocular lenses.8 Although the overall incidence of endophthalmitis among patients with foldable vs injectable lenses is not known in this study, there were twice as many endophthalmitis cases after surgery with foldable lenses than with injectable lenses. Subconjunctival antibiotics at the conclusion of cataract surgery have been associated with a lower incidence of endophthalmitis when compared with the incidence after surgery with no injected antibiotics.5,9 –11 Only two of the seven cases in the current study received subconjunctival antibiotics (Table 3, cases 1 and 6). The incidence of endophthalmitis after clear cornea phacoemulsification (0.05%) was higher than the incidence after other approaches to cataract surgery (0.02%) in the current study, but the increased incidence is not statistically significant (P ⫽ .681, Fisher’s exact test). Despite increasing utilization of clear cornea incisions and sutureless surgical technique, the incidence of endophthalmitis after cataract surgery per-
DISCUSSION CONTROVERSY EXISTS REGARDING THE POSSIBLE IN-
creased risk of postoperative endophthalmitis after cataract surgery through clear cornea incision. An experimental study of corneal wound dynamics in cadaver and rabbit corneas reported that even properly constructed corneal wounds may allow communication between the intraocular and extraocular environments.3 Some studies reported an increased risk of endophthalmitis in clear cornea cases,4,5 whereas another reported no significant difference.6 The overall incidence of endophthalmitis in the current series (0.04%) is similar to incidence rates published in recent international studies (Table 4). In a retrospective case-control study including 38 endophthalmitis patients and 371 control patients, Cooper and associates reported a threefold higher risk of endophthalmitis after cataract surgery with a clear cornea incision compared with a superior scleral tunnel incision (odds ratio, 3.36 with a 95% confidence interval).4 Among 12,317 cataract procedures, Nagaki and coworkers reported a 4.6-fold (P ⫽ .037) higher relative risk of endophthalmitis with clear cornea incisions vs superior scleral tunnel incisions.5 Colleaux and Hamilton retrospectively reviewed 13,886 cataract surgeries performed at a hospital-based surgical unit in Canada. Although the incidence of endophthalmitis was higher in clear cornea (0.129%) vs scleral tunnel (0.05%) incisions, the difference was not statistically significant.6 In a recent endophthalmitis outbreak in an Australian multi-surgeon center, 10 of 11 endophthalmitis cases followed clear cornea phacoemulsification occurring in the right eye.6 However, the surgeon’s hand dominance was 986
AMERICAN JOURNAL
OF
OPHTHALMOLOGY
JUNE 2005
formed at this institution has not significantly changed over the past 20 years.
REFERENCES 1. Aaberg TM Jr., Flynn HW Jr., Schiffman J, Newton J. Nosocomial acute-onset postoperative endophthalmitis survey. Ophthalmology 1998;105:1004 –1010. 2. Eifrig CWG, Flynn HW, Jr., Scott IU, Newton J. Acute-onset postoperative endophthalmitis: review of incidence and visual outcomes. Ophthalmic Surg Lasers 2002;33:373–378. 3. McDonnell PJ, Taban M, Sarayba M, et al. Dynamic morphology of clear corneal cataract incisions. Ophthalmology 2003;110:2342–2348. 4. Cooper BA, Holekamp NM, Bohigian G, Thompson PA. Case-control study of endophthalmitis after cataract surgery comparing scleral tunnel and clear corneal wounds. Am J Ophthalmol 2003;136:300 –305. 5. Nagaki Y, Hayasaka S, Kadoi C, et al. Bacterial endophthalmitis after small-incision cataract surgery: effect of incision placement and intraocular lens type. J Cataract Refract Surg 2003;29:20 –26. 6. Colleaux KM, Hamilton WK. Effect of prophylactic antibiotics and incision type on the incidence of endophthalmitis after cataract surgery. Can J Ophthalmol 2000;35:373–378.
VOL. 139, NO. 6
7. Maloof A, Saw V. Prophylactic intracameral vancomycin. J Cataract Refract Surg 2004;30:1610. 8. Wolner B, Liebmann JM, Sassani JW, et al. Late bleb-associated endophthalmitis after trabeculectomy with adjunctive 5-fluorouracil. Ophthalmology 1991;98:1053–1060. 9. Mayer E, Cadman D, Ewings P, et al. A 10 year retrospective survey of cataract surgery and endophthalmitis in a single eye unit: injectable lenses lower the incidence of endophthalmitis. Br J Ophthalmol 2003;87: 867– 869. 10. Buzard K, Liapis S. Prevention of endophthalmitis. J Cataract Refract Surg 2004;30:1953–1959. 11. Mandal K, Hildreth A, Farrow M, Allen D. Investigation into postoperative endophthalmitis and lessons learned. J Cataract Refract Surg 2004;30:1960 –1965. 12. Semmens JB, Li J, Morlet N, Ng J. Trends in cataract surgery and postoperative endophthalmitis in western Australia (1980 –1998): the endophthalmitis population study of western Australia. Clin Exp Ophthalmol 2003; 31:213–219. 13. Kamalarajah S, Silvestri G, Sharma N, et al. Surveillance of endophthalmitis following cataract surgery in the UK. Eye 2004;18:580 –587. 14. Wong TY, Chee SP. The epidemiology of acute endophthalmitis after cataract surgery in an asian population. Ophthalmology 2004;111:699 –705.
ENDOPHTHALMITIS FOLLOWING CATARACT SURGERY
987
Biosketch John J. Miller, MD, is currently a vitreoretinal surgical fellow at the Bascom Palmer Eye Institute, where he also completed his residency. He obtained his graduate degree from the Medical College of Georgia. His primary research interests include ocular trauma and vitreoretinal infectious diseases.
987.e1
AMERICAN JOURNAL
OF
OPHTHALMOLOGY
JUNE 2005
