INCIDENCE OF NONINFECTIOUS ENDOPHTHALMITIS AFTER INTRAVITREAL INJECTION OF PRESERVATIVE-FREE TRIAMCINOLONE ACETONIDE Daniel B. Roth, MD,*† Jonathan L. Prenner, MD,*† Orest Krajnyk, BS*
Purpose: To determine the incidence of noninfectious inflammatory endophthalmitis after preservative-free intravitreal triamcinolone acetonide (PF-IVTA) injection. Patients and Methods: We performed a retrospective interventional case series, reviewing the medical records of all patients receiving PF-IVTA injection from July 1, 2006, through December 31, 2006; 444 patients were identified who received a total of 502 PF-IVTA injections. Demographic information and details of postinjection inflammation or endophthalmitis were collected. Results: Eleven eyes (2.2%) of 11 patients (2.5%) had an inflammatory reaction after PF-IVTA injection. Complete obscuration of all fundus details was observed in seven eyes, while moderate inflammation was noted in four eyes. An inflammatory hypopyon, thought not to represent particles of triamcinolone, was seen in eight eyes. The indication for IVTA injection was cystoid macular edema in seven eyes. Conclusions: Postinjection inflammatory reactions are not uncommon after PF-IVTA injection. Noninfectious endophthalmitis after IVTA injection appears to be more common in eyes being treated for cystoid macular edema. RETINAL CASES & BRIEF REPORTS 2:247–249, 2008
From the *Department of Ophthalmology, UMDNJ– Robert Wood Johnson Medical School, New Brunswick, New Jersey; and the †Retina–Vitreous Center, P.A., New Brunswick, New Jersey.
sought to determine whether preservative-free IVTA (PF-IVTA) injection6 would reduce the incidence of this form of endophthalmitis.
T
he off-label use of intravitreal triamcinolone acetonide (IVTA) has been associated with both infectious and, more commonly, noninfectious endophthalmitis.1–5 Although noninfectious inflammatory endophthalmitis resolves spontaneously, it can pose a diagnostic challenge as to whether a true infection is actually present. In addition, the consequences of this transient intraocular inflammation are unknown. We
Methods After extensive experience with intravitreal injection of Kenalog-40 (Bristol-Myers Squibb, New York, NY), we switched to the exclusive use of preservativefree triamcinolone acetonide obtained from a commercial compounding pharmacy (New England Compounding Center, Framingham, MA), in an attempt to reduce the incidence of noninfectious endophthalmitis. The drug was prepared without any preservatives, and single-dose vials were obtained. No manipulation of the drug, removal of supernatant, filtering, or washing was performed. We reviewed the medical records of a consecutive series of patients receiving PF-IVTA
The authors have no proprietary interest in the material presented in this report. Reprint requests: Daniel B. Roth, MD, Department of Ophthalmology, UMDNJ–Robert Wood Johnson Medical School, Clinical Academic Building, 4th Floor, 125 Paterson Street, New Brunswick, NJ 08901-1977; e-mail: [email protected]
247
RETINAL CASES & BRIEF REPORTSℜ
248
●
2008
●
VOLUME 2
●
NUMBER 3
Table 1. Summary of Data for Eyes With Inflammation After PF-IVTA Injection
Patient 1 2 3 4 5 6 7 8 9 10 11
Grade of Inflammation Moderate Moderate Moderate Severe Severe Severe Severe Severe Severe Severe Severe
Hypopyon
Lens Status
History of PPV
Indication for IVTA Injection
Tap and Injection of Antibiotics
No No Yes Yes Yes Yes Yes Yes Yes Yes Yes
ACIOL PCIOL Phakic PCIOL PCIOL PCIOL PCIOL Phakic PCIOL ACIOL Phakic
Yes No No No No No No No No Yes No
Postop CME Postop CME DME Postop CME Postop CME DME DME Postop CME Postop CME CME with ERM CME/uveitis
No No Yes (NG) Yes (NG) No No No No No No No
PF-IVTA, preservative-free intravitreal triamcinolone acetonide; PPV, pars plana vitrectomy; ACIOL, anterior chamber intraocular lens; postop, postoperative; CME, cystoid macular edema; PCIOL, posterior chamber intraocular lens; DME, diabetic macular edema; NG, no growth; ERM, epiretinal membrane.
injection from July 1, 2006, through December 31, 2006. Demographic information and details of postinjection inflammation or endophthalmitis were collected. The study protocol was approved by the Institutional Review Board of the Robert Wood Johnson Medical School (New Brunswick, NJ). Results A total of 502 PF-IVTA injections were performed on 444 patients during the study period. Eleven eyes (2.2%) of 11 patients (2.5%) had an inflammatory reaction after PF-IVTA injection (Table 1). Endophthalmitis with obscuration of all fundus details was observed in seven eyes (1.4%). Moderate inflammation, defined as an anterior chamber and vitreous inflammatory reaction with a visible optic nerve, was noted in four eyes (0.8%). An inflammatory hypopyon was observed in eight eyes. Eight eyes were pseudophakic, and three eyes were phakic. In pseudophakic eyes, the posterior lens capsule was intact in four eyes, while it was open in four eyes. Three eyes had been vitrectomized before receiving PF-IVTA injection. In this cohort receiving PF-IVTA injection, the indication for IVTA injection was diabetic macular edema in 47.9%, retinal vein occlusion in 27.2%, nondiabetic cystoid macular edema in 21.3%, and age-related macular degeneration in 3.6%. In the subset of 11 eyes that developed postinjection inflammation, 7 (63.6%) were treated for nondiabetic cystoid macular edema. Nine of 11 cases resolved spontaneously without intervention. An intravitreal sample for culture was obtained from 2 of 11 eyes, and these 2 eyes received intraocular antibiotics secondary to physician concern for infectious endophthalmitis. Both of these eyes presented with severe vitritis and hypopyon
with visual acuity of hand motions. In both cases, culture yielded no growth. Discussion It has been suggested that noninfectious endophthalmitis after IVTA injection is simply pseudoendophthalmitis manifested by suspended intraocular triamcinolone particles in the absence of inflammation.7 Although this form of pseudoendophthalmitis occurs, especially in eyes with an anterior chamber intraocular lens or a posterior chamber intraocular lens with an open posterior capsule, it is a distinct clinical entity from the noninfectious inflammatory endophthalmitis that we describe in this report. Triamcinolone particles in the anterior chamber are manifested by white crystals layered in the anterior chamber angle in conjunction with free-floating particles of various sizes, usually larger than leukocytes, visible in the anterior chamber fluid. A hypopyon, inflammatory or infectious, is manifested as a cream, often fibrinoid, collection of leukocytes layered in the anterior chamber angle with leukocytes floating in the anterior chamber fluid. Although one cannot rely on this distinction with certainty without analysis of a sample from the anterior chamber, the features of pseudoendophthalmitis or triamcinolone dispersion alone can often be distinguished from inflammatory or infectious endophthalmitis by these differences. The eyes in this study had a marked anterior chamber inflammatory response, which was thought to be discernable from triamcinolone particles. In addition, significant vitritis was evident, with intraretinal hemorrhages seen in some eyes. Furthermore, this condition occurred in phakic eyes and in pseudophakic eyes with an intact posterior capsule, suggesting that the anterior chamber reaction was
249
ENDOPHTHALMITIS AFTER PF-IVTA INJECTION
not simply a collection of triamcinolone particles. The incidence of this inflammatory endophthalmitis in eyes treated for pseudophakic or uveitic CME may suggest that eyes prone to an inflammatory reaction (i.e., cystoid macular edema) are at risk for an inflammatory reaction to IVTA injection. Noninfectious endophthalmitis after IVTA injection can pose a diagnostic dilemma to clinicians. Any intraocular procedure is associated with a risk of infectious endophthalmitis, and delaying intervention with intraocular antibiotics can have devastating consequences. Therefore, if PF-IVTA injection would provide a significantly reduced incidence of postinjection inflammation, it would be preferable to Kenalog-40 injection, which is associated with a reported incidence of noninfectious endophthalmitis of 1% to 6%.1,2 Furthermore, although noninfectious endophthalmitis typically resolves spontaneously, the transient inflammation may have a deleterious effect on intraocular structures. PF-IVTA injection does not seem to eliminate the incidence of postinjection inflammatory reactions. Noninfectious endophthalmitis after IVTA injection appears to be more common in eyes being treated for postoperative cystoid macular edema and is possibly associated with the reactive inflammatory response that preexists in these eyes.
Key words: cystoid macular edema, endophthalmitis, injection, intravitreal, preservative free, triamcinolone acetonide.
References 1.
2.
3.
4.
5.
6.
7.
Nelson ML, Tennant MT, Sivalingam A, et al. Infectious and presumed noninfectious endophthalmitis after intravitreal triamcinolone acetonide injection. Retina 2003;23:686–691. Roth DB, Chieh J, Spirn MJ, et al. Noninfectious endophthalmitis associated with intravitreal triamcinolone injection. Arch Ophthalmol 2003;121:1279–1282. Moshfeghi DM, Kaiser PK, Scott IU, et al. Acute endophthalmitis following intravitreal triamcinolone acetonide injection. Am J Ophthalmol 2003;136:791–796. Sutter FK, Gillies MC. Pseudo-endophthalmitis after intravitreal injection of triamcinolone. Br J Ophthalmol 2003;87:972– 974. Moshfeghi DM, Kaiser PK, Bakri SJ, et al. Presumed sterile endophthalmitis following intravitreal triamcinolone acetonide injection. Ophthalmic Surg Lasers Imaging 2005;36: 24–29. Bakri SJ, Shah A, Falk NS, Beer PM. Intravitreal preservativefree triamcinolone acetonide for the treatment of macular oedema. Eye 2005;19:686–688. Moshfeghi AA, Scott IU, Flynn HW, Puliafito CA. Pseudohypopyon after intravitreal triamcinolone acetonide injection for cystoid macular edema. Am J Ophthalmol 2004; 138:489–492.
Purpose: To determine the incidence of noninfectious inflammatory endophthalmitis after preservative-free intravitreal triamcinolone acetonide (PF-IVTA) injection. Patients and Methods: We performed a retrospective interventional case series, reviewing the medical records of all patients receiving PF-IVTA injection from July 1, 2006, through December 31, 2006; 444 patients were identified who received a total of 502 PF-IVTA injections. Demographic information and details of postinjection inflammation or endophthalmitis were collected. Results: Eleven eyes (2.2%) of 11 patients (2.5%) had an inflammatory reaction after PF-IVTA injection. Complete obscuration of all fundus details was observed in seven eyes, while moderate inflammation was noted in four eyes. An inflammatory hypopyon, thought not to represent particles of triamcinolone, was seen in eight eyes. The indication for IVTA injection was cystoid macular edema in seven eyes. Conclusions: Postinjection inflammatory reactions are not uncommon after PF-IVTA injection. Noninfectious endophthalmitis after IVTA injection appears to be more common in eyes being treated for cystoid macular edema. RETINAL CASES & BRIEF REPORTS 2:247–249, 2008
From the *Department of Ophthalmology, UMDNJ– Robert Wood Johnson Medical School, New Brunswick, New Jersey; and the †Retina–Vitreous Center, P.A., New Brunswick, New Jersey.
sought to determine whether preservative-free IVTA (PF-IVTA) injection6 would reduce the incidence of this form of endophthalmitis.
T
he off-label use of intravitreal triamcinolone acetonide (IVTA) has been associated with both infectious and, more commonly, noninfectious endophthalmitis.1–5 Although noninfectious inflammatory endophthalmitis resolves spontaneously, it can pose a diagnostic challenge as to whether a true infection is actually present. In addition, the consequences of this transient intraocular inflammation are unknown. We
Methods After extensive experience with intravitreal injection of Kenalog-40 (Bristol-Myers Squibb, New York, NY), we switched to the exclusive use of preservativefree triamcinolone acetonide obtained from a commercial compounding pharmacy (New England Compounding Center, Framingham, MA), in an attempt to reduce the incidence of noninfectious endophthalmitis. The drug was prepared without any preservatives, and single-dose vials were obtained. No manipulation of the drug, removal of supernatant, filtering, or washing was performed. We reviewed the medical records of a consecutive series of patients receiving PF-IVTA
The authors have no proprietary interest in the material presented in this report. Reprint requests: Daniel B. Roth, MD, Department of Ophthalmology, UMDNJ–Robert Wood Johnson Medical School, Clinical Academic Building, 4th Floor, 125 Paterson Street, New Brunswick, NJ 08901-1977; e-mail: [email protected]
247
RETINAL CASES & BRIEF REPORTSℜ
248
●
2008
●
VOLUME 2
●
NUMBER 3
Table 1. Summary of Data for Eyes With Inflammation After PF-IVTA Injection
Patient 1 2 3 4 5 6 7 8 9 10 11
Grade of Inflammation Moderate Moderate Moderate Severe Severe Severe Severe Severe Severe Severe Severe
Hypopyon
Lens Status
History of PPV
Indication for IVTA Injection
Tap and Injection of Antibiotics
No No Yes Yes Yes Yes Yes Yes Yes Yes Yes
ACIOL PCIOL Phakic PCIOL PCIOL PCIOL PCIOL Phakic PCIOL ACIOL Phakic
Yes No No No No No No No No Yes No
Postop CME Postop CME DME Postop CME Postop CME DME DME Postop CME Postop CME CME with ERM CME/uveitis
No No Yes (NG) Yes (NG) No No No No No No No
PF-IVTA, preservative-free intravitreal triamcinolone acetonide; PPV, pars plana vitrectomy; ACIOL, anterior chamber intraocular lens; postop, postoperative; CME, cystoid macular edema; PCIOL, posterior chamber intraocular lens; DME, diabetic macular edema; NG, no growth; ERM, epiretinal membrane.
injection from July 1, 2006, through December 31, 2006. Demographic information and details of postinjection inflammation or endophthalmitis were collected. The study protocol was approved by the Institutional Review Board of the Robert Wood Johnson Medical School (New Brunswick, NJ). Results A total of 502 PF-IVTA injections were performed on 444 patients during the study period. Eleven eyes (2.2%) of 11 patients (2.5%) had an inflammatory reaction after PF-IVTA injection (Table 1). Endophthalmitis with obscuration of all fundus details was observed in seven eyes (1.4%). Moderate inflammation, defined as an anterior chamber and vitreous inflammatory reaction with a visible optic nerve, was noted in four eyes (0.8%). An inflammatory hypopyon was observed in eight eyes. Eight eyes were pseudophakic, and three eyes were phakic. In pseudophakic eyes, the posterior lens capsule was intact in four eyes, while it was open in four eyes. Three eyes had been vitrectomized before receiving PF-IVTA injection. In this cohort receiving PF-IVTA injection, the indication for IVTA injection was diabetic macular edema in 47.9%, retinal vein occlusion in 27.2%, nondiabetic cystoid macular edema in 21.3%, and age-related macular degeneration in 3.6%. In the subset of 11 eyes that developed postinjection inflammation, 7 (63.6%) were treated for nondiabetic cystoid macular edema. Nine of 11 cases resolved spontaneously without intervention. An intravitreal sample for culture was obtained from 2 of 11 eyes, and these 2 eyes received intraocular antibiotics secondary to physician concern for infectious endophthalmitis. Both of these eyes presented with severe vitritis and hypopyon
with visual acuity of hand motions. In both cases, culture yielded no growth. Discussion It has been suggested that noninfectious endophthalmitis after IVTA injection is simply pseudoendophthalmitis manifested by suspended intraocular triamcinolone particles in the absence of inflammation.7 Although this form of pseudoendophthalmitis occurs, especially in eyes with an anterior chamber intraocular lens or a posterior chamber intraocular lens with an open posterior capsule, it is a distinct clinical entity from the noninfectious inflammatory endophthalmitis that we describe in this report. Triamcinolone particles in the anterior chamber are manifested by white crystals layered in the anterior chamber angle in conjunction with free-floating particles of various sizes, usually larger than leukocytes, visible in the anterior chamber fluid. A hypopyon, inflammatory or infectious, is manifested as a cream, often fibrinoid, collection of leukocytes layered in the anterior chamber angle with leukocytes floating in the anterior chamber fluid. Although one cannot rely on this distinction with certainty without analysis of a sample from the anterior chamber, the features of pseudoendophthalmitis or triamcinolone dispersion alone can often be distinguished from inflammatory or infectious endophthalmitis by these differences. The eyes in this study had a marked anterior chamber inflammatory response, which was thought to be discernable from triamcinolone particles. In addition, significant vitritis was evident, with intraretinal hemorrhages seen in some eyes. Furthermore, this condition occurred in phakic eyes and in pseudophakic eyes with an intact posterior capsule, suggesting that the anterior chamber reaction was
249
ENDOPHTHALMITIS AFTER PF-IVTA INJECTION
not simply a collection of triamcinolone particles. The incidence of this inflammatory endophthalmitis in eyes treated for pseudophakic or uveitic CME may suggest that eyes prone to an inflammatory reaction (i.e., cystoid macular edema) are at risk for an inflammatory reaction to IVTA injection. Noninfectious endophthalmitis after IVTA injection can pose a diagnostic dilemma to clinicians. Any intraocular procedure is associated with a risk of infectious endophthalmitis, and delaying intervention with intraocular antibiotics can have devastating consequences. Therefore, if PF-IVTA injection would provide a significantly reduced incidence of postinjection inflammation, it would be preferable to Kenalog-40 injection, which is associated with a reported incidence of noninfectious endophthalmitis of 1% to 6%.1,2 Furthermore, although noninfectious endophthalmitis typically resolves spontaneously, the transient inflammation may have a deleterious effect on intraocular structures. PF-IVTA injection does not seem to eliminate the incidence of postinjection inflammatory reactions. Noninfectious endophthalmitis after IVTA injection appears to be more common in eyes being treated for postoperative cystoid macular edema and is possibly associated with the reactive inflammatory response that preexists in these eyes.
Key words: cystoid macular edema, endophthalmitis, injection, intravitreal, preservative free, triamcinolone acetonide.
References 1.
2.
3.
4.
5.
6.
7.
Nelson ML, Tennant MT, Sivalingam A, et al. Infectious and presumed noninfectious endophthalmitis after intravitreal triamcinolone acetonide injection. Retina 2003;23:686–691. Roth DB, Chieh J, Spirn MJ, et al. Noninfectious endophthalmitis associated with intravitreal triamcinolone injection. Arch Ophthalmol 2003;121:1279–1282. Moshfeghi DM, Kaiser PK, Scott IU, et al. Acute endophthalmitis following intravitreal triamcinolone acetonide injection. Am J Ophthalmol 2003;136:791–796. Sutter FK, Gillies MC. Pseudo-endophthalmitis after intravitreal injection of triamcinolone. Br J Ophthalmol 2003;87:972– 974. Moshfeghi DM, Kaiser PK, Bakri SJ, et al. Presumed sterile endophthalmitis following intravitreal triamcinolone acetonide injection. Ophthalmic Surg Lasers Imaging 2005;36: 24–29. Bakri SJ, Shah A, Falk NS, Beer PM. Intravitreal preservativefree triamcinolone acetonide for the treatment of macular oedema. Eye 2005;19:686–688. Moshfeghi AA, Scott IU, Flynn HW, Puliafito CA. Pseudohypopyon after intravitreal triamcinolone acetonide injection for cystoid macular edema. Am J Ophthalmol 2004; 138:489–492.
