- Email: [email protected]
Toxic anterior segment syndrome and possible association with ointment in the anterior chamber following cataract surgery
J CATARACT REFRACT SURG - VOL 32, FEBRUARY 2006
Toxic anterior segment syndrome and possible association with ointment in the anterior chamber following cataract surgery Liliana Werner, MD, PhD, Jeffrey H. Sher, MD, FRCSC, Joel R. Taylor, MD, Nick Mamalis, MD, Wesley A. Nash, MD, FRCSC, James E. Csordas, MD, CM, FRCSC, George Green, PhD, E. Peter Maziarz, PhD, X. Michael Liu, PhD
PURPOSE: To report clinical and laboratory findings of 8 cases of toxic anterior segment syndrome (TASS) related to an oily substance in the anterior chamber of patients following cataract surgery with intraocular lens (IOL) implantation. SETTING: John Moran Eye Center, University of Utah, Salt Lake City, Utah, USA. METHODS: Eight patients had uneventful phacoemulsification by the same surgeon via clear corneal incisions with implantation of the same 3-piece silicone IOL design. Postoperative medications included antibiotic/steroid ointment and pilocarpine gel; each eye was firmly patched at the end of the procedure. On the first postoperative day, some patients presented with diffuse corneal edema, increased intraocular pressure, and an oily film-like material within the anterior chamber coating the corneal endothelium. The others presented with an oily bubble floating inside the anterior chamber, which was later seen coating the IOL. Additional surgical procedures required included penetrating keratoplasty, IOL explantation, and trabeculectomy. Two corneal buttons were analyzed histopathologically. Two explanted IOLs had gross and light microscopic analyses (as well as surface analyses of 1 of them), and 4 other explanted IOLs had gas chromatography-mass spectrometry. RESULTS: Pathological examination of the corneas showed variable thinning of the epithelium with edema. The stroma was diffusely thickened and the endothelial cell layer was absent. Evaluation of the explanted IOLs confirmed the presence of an oily substance coating large areas of their anterior and posterior optic surfaces. Gas chromatography-mass spectrometry of the lens extracts identified a mixed chain hydrocarbon compound that was also found in the gas chromatography-mass spectrometry analyses of the ointment used postoperatively. CONCLUSIONS: The results indicate that the ointment gained access to the eye, causing the postoperative complications described. These cases highlight the importance of appropriate wound construction and integrity, as well as the risks of tight eye patching following placement of ointment. J Cataract Refract Surg 2006; 32:227–235 Q 2006 ASCRS and ESCRS
Toxic anterior segment syndrome (TASS) is a general term used to describe acute, sterile postoperative anterior segment reactions.1,2 Patients with TASS will often experience blurry vision with or without pain within 12 to 48 hours after surgery. Clinical findings may include diffuse corneal edema, widespread endothelial damage, anterior segment inflammation, fibrin deposits, hypopyon, dilated/irregular pupils, trabecular meshwork damage, and secondary glaucoma. It can be a major complication of cataract surgery, leading to the need for intraocular lens (IOL) exchange, secondary procedures to control the intraocular pressure Q 2006 ASCRS and ESCRS Published by Elsevier Inc.
(IOP), or penetrating keratoplasty (PKP). Known etiologies of TASS include abnormalities in intraocular irrigating solutions, preservatives in ophthalmic solutions, IOL-induced inflammation, intraocular medications, poor technique in sterilization of instruments and tubing, and particulate contamination. We describe a series of 8 cases from the same center in which TASS had different presentations after phacoemulsification with IOL implantation that could be related in each case to an oily material in the anterior chamber in the postoperative period. 0886-3350/06/$-see front matter doi:10.1016/j.jcrs.2005.12.093
227
TASS SECONDARY TO INTRAOCULAR OINTMENT
The cases consisted of 8 patients (67 to 87 years of age) operated on at the same center by the same surgeon (W.A.N.). The patients had uneventful phacoemulsification under topical/neuroleptic anesthesia between May 7, 2003, and October 21, 2003. The surgeries were performed via superior incisions with near clear corneal entry, performed with a Rhein trapezoid 2.8 to 3.2 diamond blade. These were 2-plane incisions with corneal tunnels with a length of approximately 3 mm designed to give a square configuration. A 3-piece silicone IOL (SoFlex LI 61U, Bausch & Lomb) was implanted in each case. Two patients were diabetic, and the others had no significant medical history. One hour preoperatively, lidocaine 2% gel premixed by the hospital pharmacy with cyclopentolate 1%, phenylephrine 10%, diclofenac 0.1%, and ciprofloxacin 0.3% was applied to the operative eye. Solutions used during the surgical procedure included preservative-free lidocaine 1%, balanced salt solution, epinephrine, pilocarpine hydrochloride, sodium hyaluronate 1.6% (Amvisc Plus), and vancomycin. Postoperative medications included betamethasone sodium phosphate–gentimicin sulfate ophthalmic ointment (Garasone) and pilocarpine gel. Each eye was firmly patched (single eye patch and shield) after application of the ointment at the end of the surgical procedure. A mild anterior chamber postoperative inflammatory reaction was noted in each case in the immediate postoperative period. Between the reference period (May to October 2003), the same surgeon performed 183 cataract surgery procedures using this technique.
Case Reports Case 1 The first patient was an 86-year-old woman. On the first postoperative day, cloudiness of the cornea and elevated IOP of 28 mm Hg were noted. Although the IOP could be controlled initially with acetazolamide and subsequently with timolol maleate and travoprost, diffuse corneal edema with folds in Descemet’s membrane was observed at the end of the first postoperative week (Figure 1, A). On July 15, 2003, the cornea was partially clear and the presence of an oily, film-like material in the anterior chamber extending through the pupil and coating the corneal endothelium was noted. Visual acuity at that time was counting fingers. The patient had PKP and IOL explantation/exchange. A Bausch & Lomb Surgical P366 poly(methyl methacrylate) PMMA IOL was placed in the ciliary sulcus. The corneal button and the explanted silicone IOL were sent to the John A. Moran Eye Center in 10% formalin and a balanced salt solution, respectively. The graft subsequently failed, and PKP was repeated.
Case 2 The second patient, a 75-year-old woman, had cataract surgery. The examination on the first postoperative day was unremarkable. Five days postoperatively, an oily bubble was seen floating in the anterior chamber (Figure 1, B). On July 14, 2003, the material within the bubble was found to be attached to the anterior surface of the IOL, causing a decrease in the visual acuity to 20/50. The patient had IOL explantation/exchange. A SoFlex LI 61U silicone IOL was placed in the capsular bag. The anterior chamber was extensively washed out during the procedure. The explanted silicone IOL was sent to Bausch & Lomb laboratories in the dry state. The patient’s visual acuity recovered to 20/25 postoperatively.
PATIENTS AND METHODS
Case 3
Table 1 summarizes the characteristics of the 8 patients included in this study as well as the laboratory analyses performed in each case.
The third patient, a 63-year-old woman, had cataract surgery. The examination on the first postoperative day was unremarkable. At the end of the first postoperative week, borderline elevation of the IOP (21 mm Hg) and oily debris on the anterior surface of the IOL were noted. Although steroid drops were started, increasing amounts of debris on the IOL surface as well as corneal edema were noted in the following week (Figure 1, C). The patient then had irrigation/aspiration of the anterior chamber. However, oily debris was still present in the anterior chamber after that procedure. The IOP at that time was 34 mm Hg, and the patient was treated with acetazolamide, bimatoprost, and timolol maleate/ dorzolamide. Due to the lack of improvement and persistent oily debris, a vitrectomy was performed by a vitreoretinal surgeon at another hospital. Aspirates from the anterior chamber and the vitreous were sent for cytology, revealing the presence of rare macrophages, neutrophils, and amorphous debris. This was followed by increasing corneal edema in the following months. The patient therefore had PKP and IOL explantation/exchange. A Bausch & Lomb Surgical P366 PMMA IOL was placed in the ciliary sulcus. The corneal button and the explanted silicone lens were sent to the John A. Moran Eye Center in 10% formalin and a balanced salt solution, respectively. Postoperatively, extensive peripheral anterior synechias developed, the IOP remained uncontrolled, and no light perception acuity resulted.
From the John A. Moran Eye Center, University of Utah (Werner, Taylor, Mamalis), Salt Lake City, Utah, USA, Department of Eye Medicine and Surgery of Hamilton Health Sciences (Sher, Nash, Csordas) Hamilton, Ontario, Canada, and Bausch & Lomb Inc. (Green, Maziarz, Liu) Rochester, New York, USA. Supported in part by the Research to Prevent Blindness Olga Keith Wiess Scholar Award (Dr. Werner). James P. Gilman, CRA, and Elizabeth Snodgrass, CRA, John A Moran Eye Center, assisted with the photodocumentation. Mary Mayfield, HT, John A. Moran Eye Center, provided assistance with the histopathological sections. Donggao Zhao, PhD, Electron Microscopy Center, University of South Carolina, Columbia, South Carolina, provided assistance with surface analysis. Reprint requests to Nick Mamalis, MD, John A. Moran Eye Center, University of Utah, 50 North Medical Drive, Salt Lake City, Utah 84132, USA. E-mail: [email protected].
228
J CATARACT REFRACT SURG - VOL 32, FEBRUARY 2006
TASS SECONDARY TO INTRAOCULAR OINTMENT
Table 1. Characteristics of the 8 cases included in this study.
Additional Surgical Procedures
Postoperative Clinical Presentation
Patient
Date of Surgery (Phaco C IOL Implantation)
Eye
1
05/07/03
OS
Corneal edema Elevated IOP Oily, film-like material in AC
PKP and IOL exchange (09/10/03) Second PKP (02/28/05)
2 3
05/14/03 05/28/03
OS OS
Oily bubble in AC Elevated IOP Oily debris on the IOL Corneal edema
IOL exchange (08/20/03) Washout of AC (06/18/03) Vitrectomy (07/24/03) PKP and IOL exchange (11/03/03)
4
06/25/03
OD
Oily material within pupillary area
5
07/09/03
OS
Oily bubble in AC
6
09/03/03
OD
7
10/15/03
OS
Corneal edema Oily, film-like material in AC Corneal edema Oily debris in AC
8
10/15/03
OS
Neodymium:YAG on the oily material IOL exchange (08/13/03) Aspiration of bubble and polishing of IOL (09/03/03) PKP and IOL exchange (04/26/04) Anterior vitrectomy and washout of AC (10/22/03) Trabeculectomy (11/03) Glaucoma valve (6/04) PKP and IOL exchange (01/05) Aspiration of bubble and polishing of IOL (10/22/03)
Oily bubble in AC
Laboratory Analyses Histopathology (cornea) Gross and light microscopic analyses (IOL) SEM C EDS (IOL) GC-MS (IOL) Cytology of AC and vitreous aspirates Histopathology (cornea) Gross and light microscopic analyses (IOL) GC-MS (IOL)
None GC-MS (Cornea/IOL) GC-MS (IOL)
None
AC Z anterior chamber; CG-MS Z gas chromatography-mass spectrometry; IOLZ intraocular lens; IOP Z intraocular pressure; PKP Z penetrating keratoplasty; SEM C EDS Z scanning electron microscopy C energy dispersive x-ray spectroscopy
Case 4
Case 6
The fourth patient was a 75-year-old man. On the first postoperative day, an oily material was found in the pupillary area, coating the anterior surface of the IOL. A neodymium:YAG laser was used in an attempt to disperse the oily material, but this was found to be ineffective. Intraocular lens explantation/ exchange was therefore performed. A SoFlex LI 61U silicone IOL was placed in the capsular bag. The explanted silicone lens was sent to Bausch & Lomb laboratories in the dry state. The patient’s visual acuity recovered postoperatively to 20/30C2.
The sixth patient, a 74-year-old woman, had a surgical cataract procedure. A mild corneal edema was observed on the first postoperative day, with progressive worsening in the following weeks. On November 15, 2003, an oily, film-like material was seen in the anterior chamber, coating the anterior surface of the silicone IOL. The patient had PKP and IOL explantation/ exchange. A P366 PMMA IOL was placed in the ciliary sulcus. One half of the corneal button and the explanted silicone lens were submitted to Bausch & Lomb laboratories. Case 7
Case 5 The fifth patient was an 87-year-old man. On the first postoperative day, the presence of an oily bubble in the anterior chamber was noted (Figure 1, D). The following month, the material within the bubble was found to coat the silicone IOL, causing a decrease in visual acuity (20/60). The oily bubble in the anterior chamber was aspirated. During the same procedure, a capsular polisher was used to clear the central part of the lens as much as possible of the attached oily substance. This was followed by an improvement in visual acuity to 20/30C2; the silicone IOL was not explanted.
The seventh patient, a 72-year-old woman, had cataract surgery. Mild corneal edema was observed on the first postoperative day, with progressive worsening in the subsequent weeks. Oily debris was noted in the anterior chamber on the fifth postoperative day. An anterior vitrectomy with anterior chamber washout was performed. A trabeculectomy was later performed for uncontrolled IOP elevation (42 mm Hg); the trabeculectomy failed, the IOP remained elevated (52 mm Hg), and a glaucoma valve was subsequently inserted, with appropriate glaucoma control. The patient also had PKP and IOL explantation/exchange. A
J CATARACT REFRACT SURG - VOL 32, FEBRUARY 2006
229
TASS SECONDARY TO INTRAOCULAR OINTMENT
Figure 1. Slitlamp clinical photographs from 4 of the 8 cases taken during the first postoperative week (A through D, cases 1, 2, 3, and 5, respectively). Corneal edema can be seen in A and C, where the oily material was found to be coating the corneal endothelium. In B and D, the material can be seen as a large bubble floating into the anterior chamber.
P366 PMMA IOL was placed in the ciliary sulcus. The explanted silicone lens was submitted to Bausch & Lomb laboratories. Case 8 The last patient was a 73-year-old woman. On the first postoperative day, an oily bubble was noted floating in the anterior chamber, which was subsequently coating the anterior surface of the silicone IOL. The bubble in the anterior chamber was aspirated, and the IOL anterior optic surface was cleared of the oily substance with a capsule polisher. This was followed by an improvement of visual acuity to 20/30C1; the silicone IOL was not explanted. TASS Investigation This series of cases had evaluation according to a protocol established at the John A. Moran Eye Center for TASS investigation. This involves thorough review of all aspects related to each case using a questionnaire that includes preoperative data (eg, patient information, medications), perioperative procedures (eg, operating room preparation, sterilization techniques, preoperative
230
medications, details of surgical procedure), and postoperative procedures (eg, medications, postoperative examination). From the analysis of this questionnaire, it was interesting to note that prior to the onset of these cases, a change in the operating room protocol led to the use of cut silicone tubing from the handpiece infusion line as protective sleeves for the phaco tips during the autoclaving process. Laboratory Analyses Of the 8 cases, 2 corneal buttons were sent to the John A. Moran Eye Center in 10% formalin (cases 1 and 3) and 2 explanted silicone lenses were also sent to our laboratory in contact lens vials containing balanced salt solutions (cases 1 and 3). Four other explanted IOLs (cases 2, 4, 6, and 7) and one half of a corneal button (case 6) were sent to Bausch & Lomb. One half of each corneal button analyzed in our center was processed for histopathologic evaluation. After dehydration and embedding in paraffin, serial sections were made from each specimen. The sections were then stained with hematoxylin and eosin. They were examined under a light microscope (Olympus Optical Co. Ltd.), and photomicrographs were taken for documentation.
J CATARACT REFRACT SURG - VOL 32, FEBRUARY 2006
TASS SECONDARY TO INTRAOCULAR OINTMENT
Gross examination of the explanted IOLs sent to our laboratory was performed, and gross pictures were taken using a Nikon digital camera (Model D1x with a Nikon ED 28-70 mm AF lens, Nikon). The unstained lenses were then microscopically evaluated and photographed under a light microscope (Olympus, Optical Co. Ltd.). The lens in case 1 was then sent to D. Zhao, PhD (Electron Microscopy Center, University of South Carolina). It was air dried at room temperature for at least 3 days, mounted uncoated on a carbon sticky tape on a round sample stub for imaging, and analyzed using an environmental scanning electron microscope (FEI Quanta 2000 ESEM) equipped with an energy dispersive x-ray spectroscopy detector with light element capabilities. Gas chromatography-mass spectrometry (GC-MS) was used to analyze the 4 other explanted silicone lenses sent to Baush & Lomb as well as one half of a corneal button. The specimens were removed from their containers and placed in separate glass scintillation vials. Five milliliters of tetrahydrafuran (THF) solvent was added to each vial. The specimens were allowed to extract for approximately 24 hours. Three milliliters of liquid extract was removed from each vial and placed in separate scintillation vials. The samples were concentrated by placing the extracts under a stream of nitrogen gas until all the THF solvent was removed. The residue samples were reconstituted in 70 mL of THF solvent and vortexed for several seconds. A gas-tight syringe was loaded with 2 mL of sample and injected into the GC-MS instrument (JEOL JMS-AX505HA mass spectrometer hyphenated to an HP 5890 Series II gas chromatograph). Several items from the operating room, including the silicone sleeves used for the phaco tips during the autoclaving process as well as substances used prior and immediately after the surgical procedure (eg, lidocaine gel, ointment, pilocarpine gel), were extracted and analyzed using the same GC-MS experimental conditions to determine the source of contamination. RESULTS
During the processing of the 2 corneal buttons sent to our center for histopathology, it was noted that the corneal
endothelium was loose and detaching from the tissue. Pathologic examination of both specimens showed that the corneal epithelium was variably thinned. The corneal stroma was diffusely thickened, but there was no obvious inflammatory cell reaction. Descemet’s membrane was intact, and the endothelial cell layer was completely absent (Figure 2). Gross and microscopic evaluation of the explanted lenses sent to our laboratory confirmed the presence of an oily substance coating large areas of the anterior and posterior optic surfaces of the lenses (Figure 3 and 4). Microscopically, the layer of coating material contained orange–brown pigments. Cracks formed within this layer upon drying of the specimens. Scanning electron microscopy performed on the lens in case 1 confirmed the aspect already observed on light microscopy (Figure 4, A). Energy dispersive x-ray spectroscopy performed at the level of the oily substance on the same lens showed peaks characteristic of a silicone lensdthat is, carbon (C), oxygen (O), and silicon (Si)das well as peaks of sodium (Na), chloride (Cl), potassium (K), phosphate (P), and calcium (Ca) (Figure 4, B). Gas chromatography-mass spectrometry of the lens extracts identified a mixed chain hydrocarbon compound. Figure 5, A, illustrates the total ion chromatogram from the analysis of the lens in case 4. Each peak in Figure 5, A, represents a compound eluting from the GC column. For each of these eluting peaks, a corresponding mass spectrum was obtained. The compound eluting at 14.5 minutes was identified to be a linear aliphatic hydrocarbon compound known as n-hexacosane and having the elemental composition C26H54 (Figure 5, B). The overall results identified a C21-C30 mixed chain hydrocarbon. The results from
Figure 2. Light photomicrographs of histologic sections obtained from the corneal button in case number 1 (hematoxylin and eosin stain). A: Photomicrograph showing variable areas of epithelium thinning, as well as thickening of the stroma, with condensation of posterior lamellae (original magnification 100). B: Photomicrograph showing the intact Descemet’s membrane, with complete absence of the corneal endothelium (original magnification 400).
J CATARACT REFRACT SURG - VOL 32, FEBRUARY 2006
231
TASS SECONDARY TO INTRAOCULAR OINTMENT
Figure 3. Light photomicrographs from the explanted silicone lens in case 1 (A to C). The oily material was found to coat large areas of the anterior and posterior surfaces of the lens optic, with intervening clear areas. An orange–brown pigmentation can be seen within the coating material (B and C). In the dry state, some cracks formed within the material layer (C). A: Original magnification 40. B and C: Original magnification 200.
the analyses of the lens explanted in case 2 were similar. The results of the analyses performed on the topical antibiotic/steroid ointment (Garasone) extracts matched those from the IOL extracts. Gas chromatography-mass spectrometry analysis of lidocaine gel, pilocarpine gel, and autoclaved silicone sleeves failed to match the IOL extracts. Analyses of the lenses explanted in cases 6 and 7, as well as the one half of the corneal button in case 6, also did not shown the same characteristic profile of peaks.
DISCUSSION
Given the similar profile of peaks and library match in the GC-MS analysis from 2 of 4 explanted IOLs and the ointment used postoperatively, it is reasonable to assume that the material causing the anterior segment reactions observed was the latter. After the cases reported here, the surgeon stopped using ophthalmic ointment and eye patches postoperatively and did not have any other similar case. Each gram of Garasone contains gentamicin sulfate
Figure 4. A: Scanning electron photomicrograph of the explanted lens in case 1, showing an area with a layer of the coating material. Some cracks formed within this layer in the dry state. B: Energy dispersive x-ray spectrum obtained at the level of the material coating the lens surface, showing the presence of carbon (C), oxygen (O), and silicon (Si) as well as peaks of sodium (Na), chloride (Cl), potassium (K), phosphate (P), and calcium (Ca).
232
J CATARACT REFRACT SURG - VOL 32, FEBRUARY 2006
TASS SECONDARY TO INTRAOCULAR OINTMENT
Compound eluting at 14.5 minutes: n-hexacosane 8
4
12
16
Time (minutes)
A 57
n-hexacosane
100
Figure 5. Gas chromatography-mass spectrometry analysis performed on the liquid extracts from the explanted IOL in case 4. A: Total ion chromatogram from the specimen. Each peak represents a compound eluting from the gas chromatography column. For each of these eluting peaks, a corresponding mass spectrum was obtained. B: Mass spectrum of the compound eluting at 14.5 minutes. This was identified to be a linear aliphatic hydrocarbon compound known as n-hexacosane, with the elemental composition C26H54.
71 43
85
50 99 127 366
183
0
100
200
300
400
mass (m/z)
B
(3 mg) and betamethasone sodium phosphate 1.0 mg. Nonmedicinal ingredients are mineral oil and white petrolatum. Ophthalmic ointments are typically prepared in white petrolatum, para-hydrobenzoic acids (parabens), and/or mineral oils.3 Coating of a silicone lens with an oily material can be related to silicone oil. Patients with vitreoretinal problems that may require use of silicone oil should not be implanted with silicone lenses as the oil will attach to the lens surfaces, causing optical irregularities.4 None of the patients in this series, however, had any history of vitreoretinal surgery. More recently, Ohrstrom et al.5 evaluated the silicone oil content of different ophthalmic viscosurgical devices and demonstrated that small amounts of silicone oil is a common contaminant of these solutions. However, only very small droplets of silicone oil are generally ejected from the viscosurgical devices, not near the volume of the oily substance shown in Figure 1. The effects of ointment bases in the eye were evaluated by Scheie et al.6 by injecting common ophthalmic ointment
bases in the anterior chamber of rabbit eyes. They found that the major predictor of reaction severity within the eye was the amount of ointment instilled. A volume of 0.01 cc had little or no effect on the eye, regardless of type of ointment, while a volume of 0.1 cc produced an overwhelming reaction in most eyes with prompt secondary glaucoma and loss of the globe. Of particular interest, they demonstrated that after injection of petrolatum, it migrated slowly upward and formed a translucent mass in the upper angle, becoming fixed. Localized corneal edema and haze appeared superiorly in 43% of the eyes by 2 weeks and was present in all remaining eyes by 14 weeks. Histopathologic analyses revealed absence of endothelium in the region of the cornea in contact with the petrolatum. In the cases described here, the differences in clinical presentation might have been correlated with different amounts of ointment that could penetrate the anterior chamber, although we did not have means to precisely verify that. It is noteworthy that the clinical outcomes were better when the oily material was still present as a bubble
J CATARACT REFRACT SURG - VOL 32, FEBRUARY 2006
233
TASS SECONDARY TO INTRAOCULAR OINTMENT
floating into the anterior chamber. More severe complications were observed when the oily material was already seen as a film-like material coating the corneal endothelium, before any other secondary procedure. Also, the lenses in cases 2 and 4 were explanted relatively earlier in the postoperative period and no extensive washouts of the anterior chamber were done (Table 1). These facts might have facilitated the identification of the compounds that matched the ointment used in the postoperative period by GC-MS analysis. Intraocular penetration of ointments has been described in the literature. In 1973, Fraunfelder and Hanna7 published a report on a survey sent to 400 randomly selected ophthalmologists from the Fellows in the American Academy of Ophthalmology and Otolaryngology. Of the 327 surveys returned, 65 (20%) reported having seen ointment entrapped in the anterior chamber postoperatively in a total of 95 patients. Of the 95 cases, 20 were lost to follow-up. Of the remaining 75 cases, 25 had glaucoma or uveitis thought to be a result of the ointment. Of these 25 eyes, 5 had another operation to remove the ointment and another 5 were enucleated. As it was difficult to say how many of the complications reported were directly attributable to the ointment, they performed further studies of rabbit eyes. These showed that 0.01 mL (5%) of the anterior chamber volume of most commercial ophthalmic ointment bases caused insignificant long-term reaction or damage to the eye. Therefore, the presence of a small amount of ointment entrapped in the anterior chamber may not be an indication for removal unless complications directly attributable to it are present. Garzozi and coauthors8 reported a patient who presented with a bubble floating in the anterior chamber after radial keratotomy. Ophthalmic ointment had been used after the procedure. Anterior uveitis and elevated IOP were observed during the early postoperative follow-up, as well as at 41 and 61 months postoperatively. Aralikatti et al.9 reported a patient who had uneventful phacoemulsification through an oblique, self-sealing clear corneal incision and presented with a white lump of a substance in the anterior chamber, overlying the pupil, on the first postoperative day. The surgeon assumed that the ointment used postoperatively had entered the anterior chamber and smeared the IOL. Immediate surgical removal of the ointment with IOL exchange was performed, with good outcomes. More recently, Riedl et al.10 described ointment entering the anterior chamber after cataract surgery through a temporal corneal incision. The patient was noted to have vitreous and pieces of cortex in the anterior chamber 5 weeks after surgery, followed later by inflammation and elevated IOP. By 6 months postoperatively, a 2.0 mm white, round mass was also seen swimming in the aqueous and fine droplets were seen in the anterior vitreous.
234
Chromatographic analyses of the material in the anterior chamber suggested the ointment used postoperatively was the origin of the mass. We recently evaluated the case of a patient who had uneventful phacoemulsification with implantation of a 3piece silicone IOL via a 3.0 mm scleral tunnel incision.11 Postoperative medications included antibiotic/steroid drops and ointments. Eight months postoperatively, the patient started having recurrent episodes of anterior chamber inflammatory reaction. Suspicion of lens instability causing the reactions led to 2 repositioning procedures, including performance of McCannel sutures. Finally, 18 months postoperatively, the IOL presented with a greasy film and it was later exchanged. Gas chromatography-mass spectrometry analysis of the ointment used after each surgical procedure showed several compounds that had mass spectrums characteristic of hydrocarbons similar to those detected in the extract prepared from the explanted IOL. In this case, it is possible that the ointment entered the anterior chamber after the IOL repositioning procedures, perhaps through clear corneal paracentesis usually required for the placement of McCannel iris-suture fixation.12 The first observation of globules on the IOL was only noted 5 months after the last procedure. The reasons for this late onset remain unclear to us. Chen et al.13 also recently reported a case in which an oily material was only observed in the anterior chamber in the late postoperative period after cataract surgery. The material was identified as ointment by Fourier transform infrared and confocal Raman microspectroscopies. The above-mentioned reports and the cases described here emphasize the importance of wound structure and integrity in intraocular surgery and also highlight the risks of eye patching. McDonnell et al.14 evaluated the dynamic morphology of clear corneal cataract incisions by creating clear corneal incisions in human and rabbit eyes obtained postmortem.15,16 Optical coherence tomography was used to image the corneal incisions in real time as the IOP in the eyes was varied with an infusion cannula. They found that the fluctuations in IOP within a physiologic range (!5 to 40 mm Hg) designed to mimic the pressures measured in blinking human and animal eyes after clear corneal cataract surgery caused relative movement of the wound edges. At low pressures, wound edges tended to gape, starting at the internal aspect of the wound. The same authors stated that surgeons typically examine their clear corneal incisions at the conclusion of surgery by inflating the anterior chamber with BSS, hydrating the wound to allow the swollen stroma to appose the wound edges and then applying pressure to the anterior chamber to check for leakage. However, such tests make the assumptions that the eye will remain well pressurized during the early postoperative period, that the absence of aqueous outflow from the wound correlates with the inability of surface fluid from the tear
J CATARACT REFRACT SURG - VOL 32, FEBRUARY 2006
TASS SECONDARY TO INTRAOCULAR OINTMENT
film to flow into the wound, and that the hydrated wound will not be rapidly deturgesced by the corneal endothelium. In a retrospective study, Shingleton et al.17 demonstrated that a significant percentage of eyes having clear corneal phacoemulsification had an IOP of 5 mm Hg or less 30 minutes after surgery. These facts may, in part, explain the observed higher rate of postoperative endophthalmitis, ranging from 2.6-fold to nearly 15-fold in clear corneal incisions compared to scleral incisions in cataract surgery.18–20 According to Leaming’s21 annual survey of members of the American Society of Cataract and Refractive Surgery on cataract surgery techniques, IOL preference, and contemporaneous issues, clear corneal incisions were used by 72% of respondents (up from 47% in 2000). Furthermore, the no-suture closure was preferred by 92% of respondents (86% in 2000, 95% in 2001, and 92% in 2002). Surgeons should therefore be aware of the dynamics of self-sealing clear corneal incisions as well as the associated risks of using postoperative ointments and eye patches. The possibility of intraocular penetration of any kind of ointment used postoperatively in different kinds of penetrating procedures should also be anticipated.
REFERENCES 1. Monson MC, Mamalis N, Olson RJ. Toxic anterior segment inflammation following cataract surgery. J Cataract Refract Surg 1992; 18: 184–189 2. Parikh CH, Edelhauser HF. Ocular surgical pharmacology; corneal endothelial safety and toxicity. Curr Opin Ophthalmol 2003; 14:178–185 3. Rietschel RL, Fowler JF Jr. Fisher’s Contact Dermatitis, 4th ed. Philadelphia, PA, Lippincott, Williams & Wilkins, 1995; 265–268 4. Apple DJ, Federman JL, Krolicki TJ, et al. Irreversible silicone oil adhesion to silicone intraocular lenses; a clinicopathological analysis. Ophthalmology 1996; 103:1555–1561; discussion by TM Aaberg Sr, 1561ÿ1562 5. Ohrstrom A, Svensson B, Tegenfeldt S, et al. Silicone oil content in ophthalmic viscosurgical devices. J Cataract Refract Surg 2004; 30: 1278–1280
6. Scheie HG, Rubenstein RA, Katowitz JA. Ophthalmic ointment bases in the anterior chamber; clinical and experimental observations. Arch Ophthalmol 1965; 73:36–42 7. Fraunfelder FT, Hanna C. Ophthalmic ointment. Trans Am Acad Ophthalmol Otolaryngol 1973; 77:OP467–OP475 8. Garzozi HJ, Muallem MS, Harris A. Recurrent anterior uveitis and glaucoma associated with inadvertent entry of ointment into the anterior chamber after radial keratotomy. J Cataract Refract Surg 1999; 25: 1685–1687 9. Aralikatti AKV, Needham AD, Lee MW, Prasad S. Entry of antibiotic ointment into the anterior chamber after uneventful phacoemulsification. J Cataract Refract Surg 2003; 29:595–597 10. Riedl M, Maca S, Amon M, et al. Intraocular ointment after small-incision cataract surgery causing chronic uveitis and secondary glaucoma. J Cataract Refract Surg 2003; 29:1022–1025 11. Chew J, Werner L, Mackman G, Mamalis N. Late opacification of a silicone intraocular lens due to ophthalmic ointment. J Cataract Refract Surg 2006; 32:341–346 12. Chang DF. Siepser slipknot for McCannel iris-suture fixation of subluxated intraocular lenses. J Cataract Refract Surg 2004; 30:1170–1176 13. Chen K-H, Lin S-Y, Li M-J, Cheng W-T. Retained antibiotic ophthalmic ointment on an intraocular lens 34 months after sutureless cataract surgery. Am J Ophthalmol 2005; 139:743–745 14. McDonnell PJ, Taban M, Sarayba M, et al. Dynamic morphology of clear corneal cataract incisions. Ophthalmology 2003; 110:2342–2348 15. Sarayba MA, Taban M, Ignacio TS, et al. Inflow of ocular surface fluid through clear corneal cataract incisions: a laboratory model. Am J Ophthalmol 2004; 138:206–210 16. Taban M, Rao B, Reznik J, et al. Dynamic morphology of sutureless cataract woundsdeffect of incision angle and location. Surv Ophthalmol 2004; 49(suppl 2):S62–S72 17. Shingleton BJ, Wadhwani RA, O’Donoghue MW, et al. Evaluation of intraocular pressure in the immediate period after phacoemulsification. J Cataract Refract Surg 2001; 27:524–527 18. John ME, Noblitt R. Endophthalmitis: scleral tunnel vs clear corneal incision. In: Buzard KA, Friedlander MH, Febbraro J-L, eds, The Blue Line Incision and Refractive Phacoemulsification. Thorofare, NJ, Slack, 2001; 53–56 19. 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; discussion by RA Morgan, 378 20. 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 21. Leaming DV. Practice styles and preferences of ASCRS membersd2003 survey. J Cataract Refract Surg 2004; 30:892–900
J CATARACT REFRACT SURG - VOL 32, FEBRUARY 2006
235
Toxic anterior segment syndrome and possible association with ointment in the anterior chamber following cataract surgery Liliana Werner, MD, PhD, Jeffrey H. Sher, MD, FRCSC, Joel R. Taylor, MD, Nick Mamalis, MD, Wesley A. Nash, MD, FRCSC, James E. Csordas, MD, CM, FRCSC, George Green, PhD, E. Peter Maziarz, PhD, X. Michael Liu, PhD
PURPOSE: To report clinical and laboratory findings of 8 cases of toxic anterior segment syndrome (TASS) related to an oily substance in the anterior chamber of patients following cataract surgery with intraocular lens (IOL) implantation. SETTING: John Moran Eye Center, University of Utah, Salt Lake City, Utah, USA. METHODS: Eight patients had uneventful phacoemulsification by the same surgeon via clear corneal incisions with implantation of the same 3-piece silicone IOL design. Postoperative medications included antibiotic/steroid ointment and pilocarpine gel; each eye was firmly patched at the end of the procedure. On the first postoperative day, some patients presented with diffuse corneal edema, increased intraocular pressure, and an oily film-like material within the anterior chamber coating the corneal endothelium. The others presented with an oily bubble floating inside the anterior chamber, which was later seen coating the IOL. Additional surgical procedures required included penetrating keratoplasty, IOL explantation, and trabeculectomy. Two corneal buttons were analyzed histopathologically. Two explanted IOLs had gross and light microscopic analyses (as well as surface analyses of 1 of them), and 4 other explanted IOLs had gas chromatography-mass spectrometry. RESULTS: Pathological examination of the corneas showed variable thinning of the epithelium with edema. The stroma was diffusely thickened and the endothelial cell layer was absent. Evaluation of the explanted IOLs confirmed the presence of an oily substance coating large areas of their anterior and posterior optic surfaces. Gas chromatography-mass spectrometry of the lens extracts identified a mixed chain hydrocarbon compound that was also found in the gas chromatography-mass spectrometry analyses of the ointment used postoperatively. CONCLUSIONS: The results indicate that the ointment gained access to the eye, causing the postoperative complications described. These cases highlight the importance of appropriate wound construction and integrity, as well as the risks of tight eye patching following placement of ointment. J Cataract Refract Surg 2006; 32:227–235 Q 2006 ASCRS and ESCRS
Toxic anterior segment syndrome (TASS) is a general term used to describe acute, sterile postoperative anterior segment reactions.1,2 Patients with TASS will often experience blurry vision with or without pain within 12 to 48 hours after surgery. Clinical findings may include diffuse corneal edema, widespread endothelial damage, anterior segment inflammation, fibrin deposits, hypopyon, dilated/irregular pupils, trabecular meshwork damage, and secondary glaucoma. It can be a major complication of cataract surgery, leading to the need for intraocular lens (IOL) exchange, secondary procedures to control the intraocular pressure Q 2006 ASCRS and ESCRS Published by Elsevier Inc.
(IOP), or penetrating keratoplasty (PKP). Known etiologies of TASS include abnormalities in intraocular irrigating solutions, preservatives in ophthalmic solutions, IOL-induced inflammation, intraocular medications, poor technique in sterilization of instruments and tubing, and particulate contamination. We describe a series of 8 cases from the same center in which TASS had different presentations after phacoemulsification with IOL implantation that could be related in each case to an oily material in the anterior chamber in the postoperative period. 0886-3350/06/$-see front matter doi:10.1016/j.jcrs.2005.12.093
227
TASS SECONDARY TO INTRAOCULAR OINTMENT
The cases consisted of 8 patients (67 to 87 years of age) operated on at the same center by the same surgeon (W.A.N.). The patients had uneventful phacoemulsification under topical/neuroleptic anesthesia between May 7, 2003, and October 21, 2003. The surgeries were performed via superior incisions with near clear corneal entry, performed with a Rhein trapezoid 2.8 to 3.2 diamond blade. These were 2-plane incisions with corneal tunnels with a length of approximately 3 mm designed to give a square configuration. A 3-piece silicone IOL (SoFlex LI 61U, Bausch & Lomb) was implanted in each case. Two patients were diabetic, and the others had no significant medical history. One hour preoperatively, lidocaine 2% gel premixed by the hospital pharmacy with cyclopentolate 1%, phenylephrine 10%, diclofenac 0.1%, and ciprofloxacin 0.3% was applied to the operative eye. Solutions used during the surgical procedure included preservative-free lidocaine 1%, balanced salt solution, epinephrine, pilocarpine hydrochloride, sodium hyaluronate 1.6% (Amvisc Plus), and vancomycin. Postoperative medications included betamethasone sodium phosphate–gentimicin sulfate ophthalmic ointment (Garasone) and pilocarpine gel. Each eye was firmly patched (single eye patch and shield) after application of the ointment at the end of the surgical procedure. A mild anterior chamber postoperative inflammatory reaction was noted in each case in the immediate postoperative period. Between the reference period (May to October 2003), the same surgeon performed 183 cataract surgery procedures using this technique.
Case Reports Case 1 The first patient was an 86-year-old woman. On the first postoperative day, cloudiness of the cornea and elevated IOP of 28 mm Hg were noted. Although the IOP could be controlled initially with acetazolamide and subsequently with timolol maleate and travoprost, diffuse corneal edema with folds in Descemet’s membrane was observed at the end of the first postoperative week (Figure 1, A). On July 15, 2003, the cornea was partially clear and the presence of an oily, film-like material in the anterior chamber extending through the pupil and coating the corneal endothelium was noted. Visual acuity at that time was counting fingers. The patient had PKP and IOL explantation/exchange. A Bausch & Lomb Surgical P366 poly(methyl methacrylate) PMMA IOL was placed in the ciliary sulcus. The corneal button and the explanted silicone IOL were sent to the John A. Moran Eye Center in 10% formalin and a balanced salt solution, respectively. The graft subsequently failed, and PKP was repeated.
Case 2 The second patient, a 75-year-old woman, had cataract surgery. The examination on the first postoperative day was unremarkable. Five days postoperatively, an oily bubble was seen floating in the anterior chamber (Figure 1, B). On July 14, 2003, the material within the bubble was found to be attached to the anterior surface of the IOL, causing a decrease in the visual acuity to 20/50. The patient had IOL explantation/exchange. A SoFlex LI 61U silicone IOL was placed in the capsular bag. The anterior chamber was extensively washed out during the procedure. The explanted silicone IOL was sent to Bausch & Lomb laboratories in the dry state. The patient’s visual acuity recovered to 20/25 postoperatively.
PATIENTS AND METHODS
Case 3
Table 1 summarizes the characteristics of the 8 patients included in this study as well as the laboratory analyses performed in each case.
The third patient, a 63-year-old woman, had cataract surgery. The examination on the first postoperative day was unremarkable. At the end of the first postoperative week, borderline elevation of the IOP (21 mm Hg) and oily debris on the anterior surface of the IOL were noted. Although steroid drops were started, increasing amounts of debris on the IOL surface as well as corneal edema were noted in the following week (Figure 1, C). The patient then had irrigation/aspiration of the anterior chamber. However, oily debris was still present in the anterior chamber after that procedure. The IOP at that time was 34 mm Hg, and the patient was treated with acetazolamide, bimatoprost, and timolol maleate/ dorzolamide. Due to the lack of improvement and persistent oily debris, a vitrectomy was performed by a vitreoretinal surgeon at another hospital. Aspirates from the anterior chamber and the vitreous were sent for cytology, revealing the presence of rare macrophages, neutrophils, and amorphous debris. This was followed by increasing corneal edema in the following months. The patient therefore had PKP and IOL explantation/exchange. A Bausch & Lomb Surgical P366 PMMA IOL was placed in the ciliary sulcus. The corneal button and the explanted silicone lens were sent to the John A. Moran Eye Center in 10% formalin and a balanced salt solution, respectively. Postoperatively, extensive peripheral anterior synechias developed, the IOP remained uncontrolled, and no light perception acuity resulted.
From the John A. Moran Eye Center, University of Utah (Werner, Taylor, Mamalis), Salt Lake City, Utah, USA, Department of Eye Medicine and Surgery of Hamilton Health Sciences (Sher, Nash, Csordas) Hamilton, Ontario, Canada, and Bausch & Lomb Inc. (Green, Maziarz, Liu) Rochester, New York, USA. Supported in part by the Research to Prevent Blindness Olga Keith Wiess Scholar Award (Dr. Werner). James P. Gilman, CRA, and Elizabeth Snodgrass, CRA, John A Moran Eye Center, assisted with the photodocumentation. Mary Mayfield, HT, John A. Moran Eye Center, provided assistance with the histopathological sections. Donggao Zhao, PhD, Electron Microscopy Center, University of South Carolina, Columbia, South Carolina, provided assistance with surface analysis. Reprint requests to Nick Mamalis, MD, John A. Moran Eye Center, University of Utah, 50 North Medical Drive, Salt Lake City, Utah 84132, USA. E-mail: [email protected].
228
J CATARACT REFRACT SURG - VOL 32, FEBRUARY 2006
TASS SECONDARY TO INTRAOCULAR OINTMENT
Table 1. Characteristics of the 8 cases included in this study.
Additional Surgical Procedures
Postoperative Clinical Presentation
Patient
Date of Surgery (Phaco C IOL Implantation)
Eye
1
05/07/03
OS
Corneal edema Elevated IOP Oily, film-like material in AC
PKP and IOL exchange (09/10/03) Second PKP (02/28/05)
2 3
05/14/03 05/28/03
OS OS
Oily bubble in AC Elevated IOP Oily debris on the IOL Corneal edema
IOL exchange (08/20/03) Washout of AC (06/18/03) Vitrectomy (07/24/03) PKP and IOL exchange (11/03/03)
4
06/25/03
OD
Oily material within pupillary area
5
07/09/03
OS
Oily bubble in AC
6
09/03/03
OD
7
10/15/03
OS
Corneal edema Oily, film-like material in AC Corneal edema Oily debris in AC
8
10/15/03
OS
Neodymium:YAG on the oily material IOL exchange (08/13/03) Aspiration of bubble and polishing of IOL (09/03/03) PKP and IOL exchange (04/26/04) Anterior vitrectomy and washout of AC (10/22/03) Trabeculectomy (11/03) Glaucoma valve (6/04) PKP and IOL exchange (01/05) Aspiration of bubble and polishing of IOL (10/22/03)
Oily bubble in AC
Laboratory Analyses Histopathology (cornea) Gross and light microscopic analyses (IOL) SEM C EDS (IOL) GC-MS (IOL) Cytology of AC and vitreous aspirates Histopathology (cornea) Gross and light microscopic analyses (IOL) GC-MS (IOL)
None GC-MS (Cornea/IOL) GC-MS (IOL)
None
AC Z anterior chamber; CG-MS Z gas chromatography-mass spectrometry; IOLZ intraocular lens; IOP Z intraocular pressure; PKP Z penetrating keratoplasty; SEM C EDS Z scanning electron microscopy C energy dispersive x-ray spectroscopy
Case 4
Case 6
The fourth patient was a 75-year-old man. On the first postoperative day, an oily material was found in the pupillary area, coating the anterior surface of the IOL. A neodymium:YAG laser was used in an attempt to disperse the oily material, but this was found to be ineffective. Intraocular lens explantation/ exchange was therefore performed. A SoFlex LI 61U silicone IOL was placed in the capsular bag. The explanted silicone lens was sent to Bausch & Lomb laboratories in the dry state. The patient’s visual acuity recovered postoperatively to 20/30C2.
The sixth patient, a 74-year-old woman, had a surgical cataract procedure. A mild corneal edema was observed on the first postoperative day, with progressive worsening in the following weeks. On November 15, 2003, an oily, film-like material was seen in the anterior chamber, coating the anterior surface of the silicone IOL. The patient had PKP and IOL explantation/ exchange. A P366 PMMA IOL was placed in the ciliary sulcus. One half of the corneal button and the explanted silicone lens were submitted to Bausch & Lomb laboratories. Case 7
Case 5 The fifth patient was an 87-year-old man. On the first postoperative day, the presence of an oily bubble in the anterior chamber was noted (Figure 1, D). The following month, the material within the bubble was found to coat the silicone IOL, causing a decrease in visual acuity (20/60). The oily bubble in the anterior chamber was aspirated. During the same procedure, a capsular polisher was used to clear the central part of the lens as much as possible of the attached oily substance. This was followed by an improvement in visual acuity to 20/30C2; the silicone IOL was not explanted.
The seventh patient, a 72-year-old woman, had cataract surgery. Mild corneal edema was observed on the first postoperative day, with progressive worsening in the subsequent weeks. Oily debris was noted in the anterior chamber on the fifth postoperative day. An anterior vitrectomy with anterior chamber washout was performed. A trabeculectomy was later performed for uncontrolled IOP elevation (42 mm Hg); the trabeculectomy failed, the IOP remained elevated (52 mm Hg), and a glaucoma valve was subsequently inserted, with appropriate glaucoma control. The patient also had PKP and IOL explantation/exchange. A
J CATARACT REFRACT SURG - VOL 32, FEBRUARY 2006
229
TASS SECONDARY TO INTRAOCULAR OINTMENT
Figure 1. Slitlamp clinical photographs from 4 of the 8 cases taken during the first postoperative week (A through D, cases 1, 2, 3, and 5, respectively). Corneal edema can be seen in A and C, where the oily material was found to be coating the corneal endothelium. In B and D, the material can be seen as a large bubble floating into the anterior chamber.
P366 PMMA IOL was placed in the ciliary sulcus. The explanted silicone lens was submitted to Bausch & Lomb laboratories. Case 8 The last patient was a 73-year-old woman. On the first postoperative day, an oily bubble was noted floating in the anterior chamber, which was subsequently coating the anterior surface of the silicone IOL. The bubble in the anterior chamber was aspirated, and the IOL anterior optic surface was cleared of the oily substance with a capsule polisher. This was followed by an improvement of visual acuity to 20/30C1; the silicone IOL was not explanted. TASS Investigation This series of cases had evaluation according to a protocol established at the John A. Moran Eye Center for TASS investigation. This involves thorough review of all aspects related to each case using a questionnaire that includes preoperative data (eg, patient information, medications), perioperative procedures (eg, operating room preparation, sterilization techniques, preoperative
230
medications, details of surgical procedure), and postoperative procedures (eg, medications, postoperative examination). From the analysis of this questionnaire, it was interesting to note that prior to the onset of these cases, a change in the operating room protocol led to the use of cut silicone tubing from the handpiece infusion line as protective sleeves for the phaco tips during the autoclaving process. Laboratory Analyses Of the 8 cases, 2 corneal buttons were sent to the John A. Moran Eye Center in 10% formalin (cases 1 and 3) and 2 explanted silicone lenses were also sent to our laboratory in contact lens vials containing balanced salt solutions (cases 1 and 3). Four other explanted IOLs (cases 2, 4, 6, and 7) and one half of a corneal button (case 6) were sent to Bausch & Lomb. One half of each corneal button analyzed in our center was processed for histopathologic evaluation. After dehydration and embedding in paraffin, serial sections were made from each specimen. The sections were then stained with hematoxylin and eosin. They were examined under a light microscope (Olympus Optical Co. Ltd.), and photomicrographs were taken for documentation.
J CATARACT REFRACT SURG - VOL 32, FEBRUARY 2006
TASS SECONDARY TO INTRAOCULAR OINTMENT
Gross examination of the explanted IOLs sent to our laboratory was performed, and gross pictures were taken using a Nikon digital camera (Model D1x with a Nikon ED 28-70 mm AF lens, Nikon). The unstained lenses were then microscopically evaluated and photographed under a light microscope (Olympus, Optical Co. Ltd.). The lens in case 1 was then sent to D. Zhao, PhD (Electron Microscopy Center, University of South Carolina). It was air dried at room temperature for at least 3 days, mounted uncoated on a carbon sticky tape on a round sample stub for imaging, and analyzed using an environmental scanning electron microscope (FEI Quanta 2000 ESEM) equipped with an energy dispersive x-ray spectroscopy detector with light element capabilities. Gas chromatography-mass spectrometry (GC-MS) was used to analyze the 4 other explanted silicone lenses sent to Baush & Lomb as well as one half of a corneal button. The specimens were removed from their containers and placed in separate glass scintillation vials. Five milliliters of tetrahydrafuran (THF) solvent was added to each vial. The specimens were allowed to extract for approximately 24 hours. Three milliliters of liquid extract was removed from each vial and placed in separate scintillation vials. The samples were concentrated by placing the extracts under a stream of nitrogen gas until all the THF solvent was removed. The residue samples were reconstituted in 70 mL of THF solvent and vortexed for several seconds. A gas-tight syringe was loaded with 2 mL of sample and injected into the GC-MS instrument (JEOL JMS-AX505HA mass spectrometer hyphenated to an HP 5890 Series II gas chromatograph). Several items from the operating room, including the silicone sleeves used for the phaco tips during the autoclaving process as well as substances used prior and immediately after the surgical procedure (eg, lidocaine gel, ointment, pilocarpine gel), were extracted and analyzed using the same GC-MS experimental conditions to determine the source of contamination. RESULTS
During the processing of the 2 corneal buttons sent to our center for histopathology, it was noted that the corneal
endothelium was loose and detaching from the tissue. Pathologic examination of both specimens showed that the corneal epithelium was variably thinned. The corneal stroma was diffusely thickened, but there was no obvious inflammatory cell reaction. Descemet’s membrane was intact, and the endothelial cell layer was completely absent (Figure 2). Gross and microscopic evaluation of the explanted lenses sent to our laboratory confirmed the presence of an oily substance coating large areas of the anterior and posterior optic surfaces of the lenses (Figure 3 and 4). Microscopically, the layer of coating material contained orange–brown pigments. Cracks formed within this layer upon drying of the specimens. Scanning electron microscopy performed on the lens in case 1 confirmed the aspect already observed on light microscopy (Figure 4, A). Energy dispersive x-ray spectroscopy performed at the level of the oily substance on the same lens showed peaks characteristic of a silicone lensdthat is, carbon (C), oxygen (O), and silicon (Si)das well as peaks of sodium (Na), chloride (Cl), potassium (K), phosphate (P), and calcium (Ca) (Figure 4, B). Gas chromatography-mass spectrometry of the lens extracts identified a mixed chain hydrocarbon compound. Figure 5, A, illustrates the total ion chromatogram from the analysis of the lens in case 4. Each peak in Figure 5, A, represents a compound eluting from the GC column. For each of these eluting peaks, a corresponding mass spectrum was obtained. The compound eluting at 14.5 minutes was identified to be a linear aliphatic hydrocarbon compound known as n-hexacosane and having the elemental composition C26H54 (Figure 5, B). The overall results identified a C21-C30 mixed chain hydrocarbon. The results from
Figure 2. Light photomicrographs of histologic sections obtained from the corneal button in case number 1 (hematoxylin and eosin stain). A: Photomicrograph showing variable areas of epithelium thinning, as well as thickening of the stroma, with condensation of posterior lamellae (original magnification 100). B: Photomicrograph showing the intact Descemet’s membrane, with complete absence of the corneal endothelium (original magnification 400).
J CATARACT REFRACT SURG - VOL 32, FEBRUARY 2006
231
TASS SECONDARY TO INTRAOCULAR OINTMENT
Figure 3. Light photomicrographs from the explanted silicone lens in case 1 (A to C). The oily material was found to coat large areas of the anterior and posterior surfaces of the lens optic, with intervening clear areas. An orange–brown pigmentation can be seen within the coating material (B and C). In the dry state, some cracks formed within the material layer (C). A: Original magnification 40. B and C: Original magnification 200.
the analyses of the lens explanted in case 2 were similar. The results of the analyses performed on the topical antibiotic/steroid ointment (Garasone) extracts matched those from the IOL extracts. Gas chromatography-mass spectrometry analysis of lidocaine gel, pilocarpine gel, and autoclaved silicone sleeves failed to match the IOL extracts. Analyses of the lenses explanted in cases 6 and 7, as well as the one half of the corneal button in case 6, also did not shown the same characteristic profile of peaks.
DISCUSSION
Given the similar profile of peaks and library match in the GC-MS analysis from 2 of 4 explanted IOLs and the ointment used postoperatively, it is reasonable to assume that the material causing the anterior segment reactions observed was the latter. After the cases reported here, the surgeon stopped using ophthalmic ointment and eye patches postoperatively and did not have any other similar case. Each gram of Garasone contains gentamicin sulfate
Figure 4. A: Scanning electron photomicrograph of the explanted lens in case 1, showing an area with a layer of the coating material. Some cracks formed within this layer in the dry state. B: Energy dispersive x-ray spectrum obtained at the level of the material coating the lens surface, showing the presence of carbon (C), oxygen (O), and silicon (Si) as well as peaks of sodium (Na), chloride (Cl), potassium (K), phosphate (P), and calcium (Ca).
232
J CATARACT REFRACT SURG - VOL 32, FEBRUARY 2006
TASS SECONDARY TO INTRAOCULAR OINTMENT
Compound eluting at 14.5 minutes: n-hexacosane 8
4
12
16
Time (minutes)
A 57
n-hexacosane
100
Figure 5. Gas chromatography-mass spectrometry analysis performed on the liquid extracts from the explanted IOL in case 4. A: Total ion chromatogram from the specimen. Each peak represents a compound eluting from the gas chromatography column. For each of these eluting peaks, a corresponding mass spectrum was obtained. B: Mass spectrum of the compound eluting at 14.5 minutes. This was identified to be a linear aliphatic hydrocarbon compound known as n-hexacosane, with the elemental composition C26H54.
71 43
85
50 99 127 366
183
0
100
200
300
400
mass (m/z)
B
(3 mg) and betamethasone sodium phosphate 1.0 mg. Nonmedicinal ingredients are mineral oil and white petrolatum. Ophthalmic ointments are typically prepared in white petrolatum, para-hydrobenzoic acids (parabens), and/or mineral oils.3 Coating of a silicone lens with an oily material can be related to silicone oil. Patients with vitreoretinal problems that may require use of silicone oil should not be implanted with silicone lenses as the oil will attach to the lens surfaces, causing optical irregularities.4 None of the patients in this series, however, had any history of vitreoretinal surgery. More recently, Ohrstrom et al.5 evaluated the silicone oil content of different ophthalmic viscosurgical devices and demonstrated that small amounts of silicone oil is a common contaminant of these solutions. However, only very small droplets of silicone oil are generally ejected from the viscosurgical devices, not near the volume of the oily substance shown in Figure 1. The effects of ointment bases in the eye were evaluated by Scheie et al.6 by injecting common ophthalmic ointment
bases in the anterior chamber of rabbit eyes. They found that the major predictor of reaction severity within the eye was the amount of ointment instilled. A volume of 0.01 cc had little or no effect on the eye, regardless of type of ointment, while a volume of 0.1 cc produced an overwhelming reaction in most eyes with prompt secondary glaucoma and loss of the globe. Of particular interest, they demonstrated that after injection of petrolatum, it migrated slowly upward and formed a translucent mass in the upper angle, becoming fixed. Localized corneal edema and haze appeared superiorly in 43% of the eyes by 2 weeks and was present in all remaining eyes by 14 weeks. Histopathologic analyses revealed absence of endothelium in the region of the cornea in contact with the petrolatum. In the cases described here, the differences in clinical presentation might have been correlated with different amounts of ointment that could penetrate the anterior chamber, although we did not have means to precisely verify that. It is noteworthy that the clinical outcomes were better when the oily material was still present as a bubble
J CATARACT REFRACT SURG - VOL 32, FEBRUARY 2006
233
TASS SECONDARY TO INTRAOCULAR OINTMENT
floating into the anterior chamber. More severe complications were observed when the oily material was already seen as a film-like material coating the corneal endothelium, before any other secondary procedure. Also, the lenses in cases 2 and 4 were explanted relatively earlier in the postoperative period and no extensive washouts of the anterior chamber were done (Table 1). These facts might have facilitated the identification of the compounds that matched the ointment used in the postoperative period by GC-MS analysis. Intraocular penetration of ointments has been described in the literature. In 1973, Fraunfelder and Hanna7 published a report on a survey sent to 400 randomly selected ophthalmologists from the Fellows in the American Academy of Ophthalmology and Otolaryngology. Of the 327 surveys returned, 65 (20%) reported having seen ointment entrapped in the anterior chamber postoperatively in a total of 95 patients. Of the 95 cases, 20 were lost to follow-up. Of the remaining 75 cases, 25 had glaucoma or uveitis thought to be a result of the ointment. Of these 25 eyes, 5 had another operation to remove the ointment and another 5 were enucleated. As it was difficult to say how many of the complications reported were directly attributable to the ointment, they performed further studies of rabbit eyes. These showed that 0.01 mL (5%) of the anterior chamber volume of most commercial ophthalmic ointment bases caused insignificant long-term reaction or damage to the eye. Therefore, the presence of a small amount of ointment entrapped in the anterior chamber may not be an indication for removal unless complications directly attributable to it are present. Garzozi and coauthors8 reported a patient who presented with a bubble floating in the anterior chamber after radial keratotomy. Ophthalmic ointment had been used after the procedure. Anterior uveitis and elevated IOP were observed during the early postoperative follow-up, as well as at 41 and 61 months postoperatively. Aralikatti et al.9 reported a patient who had uneventful phacoemulsification through an oblique, self-sealing clear corneal incision and presented with a white lump of a substance in the anterior chamber, overlying the pupil, on the first postoperative day. The surgeon assumed that the ointment used postoperatively had entered the anterior chamber and smeared the IOL. Immediate surgical removal of the ointment with IOL exchange was performed, with good outcomes. More recently, Riedl et al.10 described ointment entering the anterior chamber after cataract surgery through a temporal corneal incision. The patient was noted to have vitreous and pieces of cortex in the anterior chamber 5 weeks after surgery, followed later by inflammation and elevated IOP. By 6 months postoperatively, a 2.0 mm white, round mass was also seen swimming in the aqueous and fine droplets were seen in the anterior vitreous.
234
Chromatographic analyses of the material in the anterior chamber suggested the ointment used postoperatively was the origin of the mass. We recently evaluated the case of a patient who had uneventful phacoemulsification with implantation of a 3piece silicone IOL via a 3.0 mm scleral tunnel incision.11 Postoperative medications included antibiotic/steroid drops and ointments. Eight months postoperatively, the patient started having recurrent episodes of anterior chamber inflammatory reaction. Suspicion of lens instability causing the reactions led to 2 repositioning procedures, including performance of McCannel sutures. Finally, 18 months postoperatively, the IOL presented with a greasy film and it was later exchanged. Gas chromatography-mass spectrometry analysis of the ointment used after each surgical procedure showed several compounds that had mass spectrums characteristic of hydrocarbons similar to those detected in the extract prepared from the explanted IOL. In this case, it is possible that the ointment entered the anterior chamber after the IOL repositioning procedures, perhaps through clear corneal paracentesis usually required for the placement of McCannel iris-suture fixation.12 The first observation of globules on the IOL was only noted 5 months after the last procedure. The reasons for this late onset remain unclear to us. Chen et al.13 also recently reported a case in which an oily material was only observed in the anterior chamber in the late postoperative period after cataract surgery. The material was identified as ointment by Fourier transform infrared and confocal Raman microspectroscopies. The above-mentioned reports and the cases described here emphasize the importance of wound structure and integrity in intraocular surgery and also highlight the risks of eye patching. McDonnell et al.14 evaluated the dynamic morphology of clear corneal cataract incisions by creating clear corneal incisions in human and rabbit eyes obtained postmortem.15,16 Optical coherence tomography was used to image the corneal incisions in real time as the IOP in the eyes was varied with an infusion cannula. They found that the fluctuations in IOP within a physiologic range (!5 to 40 mm Hg) designed to mimic the pressures measured in blinking human and animal eyes after clear corneal cataract surgery caused relative movement of the wound edges. At low pressures, wound edges tended to gape, starting at the internal aspect of the wound. The same authors stated that surgeons typically examine their clear corneal incisions at the conclusion of surgery by inflating the anterior chamber with BSS, hydrating the wound to allow the swollen stroma to appose the wound edges and then applying pressure to the anterior chamber to check for leakage. However, such tests make the assumptions that the eye will remain well pressurized during the early postoperative period, that the absence of aqueous outflow from the wound correlates with the inability of surface fluid from the tear
J CATARACT REFRACT SURG - VOL 32, FEBRUARY 2006
TASS SECONDARY TO INTRAOCULAR OINTMENT
film to flow into the wound, and that the hydrated wound will not be rapidly deturgesced by the corneal endothelium. In a retrospective study, Shingleton et al.17 demonstrated that a significant percentage of eyes having clear corneal phacoemulsification had an IOP of 5 mm Hg or less 30 minutes after surgery. These facts may, in part, explain the observed higher rate of postoperative endophthalmitis, ranging from 2.6-fold to nearly 15-fold in clear corneal incisions compared to scleral incisions in cataract surgery.18–20 According to Leaming’s21 annual survey of members of the American Society of Cataract and Refractive Surgery on cataract surgery techniques, IOL preference, and contemporaneous issues, clear corneal incisions were used by 72% of respondents (up from 47% in 2000). Furthermore, the no-suture closure was preferred by 92% of respondents (86% in 2000, 95% in 2001, and 92% in 2002). Surgeons should therefore be aware of the dynamics of self-sealing clear corneal incisions as well as the associated risks of using postoperative ointments and eye patches. The possibility of intraocular penetration of any kind of ointment used postoperatively in different kinds of penetrating procedures should also be anticipated.
REFERENCES 1. Monson MC, Mamalis N, Olson RJ. Toxic anterior segment inflammation following cataract surgery. J Cataract Refract Surg 1992; 18: 184–189 2. Parikh CH, Edelhauser HF. Ocular surgical pharmacology; corneal endothelial safety and toxicity. Curr Opin Ophthalmol 2003; 14:178–185 3. Rietschel RL, Fowler JF Jr. Fisher’s Contact Dermatitis, 4th ed. Philadelphia, PA, Lippincott, Williams & Wilkins, 1995; 265–268 4. Apple DJ, Federman JL, Krolicki TJ, et al. Irreversible silicone oil adhesion to silicone intraocular lenses; a clinicopathological analysis. Ophthalmology 1996; 103:1555–1561; discussion by TM Aaberg Sr, 1561ÿ1562 5. Ohrstrom A, Svensson B, Tegenfeldt S, et al. Silicone oil content in ophthalmic viscosurgical devices. J Cataract Refract Surg 2004; 30: 1278–1280
6. Scheie HG, Rubenstein RA, Katowitz JA. Ophthalmic ointment bases in the anterior chamber; clinical and experimental observations. Arch Ophthalmol 1965; 73:36–42 7. Fraunfelder FT, Hanna C. Ophthalmic ointment. Trans Am Acad Ophthalmol Otolaryngol 1973; 77:OP467–OP475 8. Garzozi HJ, Muallem MS, Harris A. Recurrent anterior uveitis and glaucoma associated with inadvertent entry of ointment into the anterior chamber after radial keratotomy. J Cataract Refract Surg 1999; 25: 1685–1687 9. Aralikatti AKV, Needham AD, Lee MW, Prasad S. Entry of antibiotic ointment into the anterior chamber after uneventful phacoemulsification. J Cataract Refract Surg 2003; 29:595–597 10. Riedl M, Maca S, Amon M, et al. Intraocular ointment after small-incision cataract surgery causing chronic uveitis and secondary glaucoma. J Cataract Refract Surg 2003; 29:1022–1025 11. Chew J, Werner L, Mackman G, Mamalis N. Late opacification of a silicone intraocular lens due to ophthalmic ointment. J Cataract Refract Surg 2006; 32:341–346 12. Chang DF. Siepser slipknot for McCannel iris-suture fixation of subluxated intraocular lenses. J Cataract Refract Surg 2004; 30:1170–1176 13. Chen K-H, Lin S-Y, Li M-J, Cheng W-T. Retained antibiotic ophthalmic ointment on an intraocular lens 34 months after sutureless cataract surgery. Am J Ophthalmol 2005; 139:743–745 14. McDonnell PJ, Taban M, Sarayba M, et al. Dynamic morphology of clear corneal cataract incisions. Ophthalmology 2003; 110:2342–2348 15. Sarayba MA, Taban M, Ignacio TS, et al. Inflow of ocular surface fluid through clear corneal cataract incisions: a laboratory model. Am J Ophthalmol 2004; 138:206–210 16. Taban M, Rao B, Reznik J, et al. Dynamic morphology of sutureless cataract woundsdeffect of incision angle and location. Surv Ophthalmol 2004; 49(suppl 2):S62–S72 17. Shingleton BJ, Wadhwani RA, O’Donoghue MW, et al. Evaluation of intraocular pressure in the immediate period after phacoemulsification. J Cataract Refract Surg 2001; 27:524–527 18. John ME, Noblitt R. Endophthalmitis: scleral tunnel vs clear corneal incision. In: Buzard KA, Friedlander MH, Febbraro J-L, eds, The Blue Line Incision and Refractive Phacoemulsification. Thorofare, NJ, Slack, 2001; 53–56 19. 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; discussion by RA Morgan, 378 20. 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 21. Leaming DV. Practice styles and preferences of ASCRS membersd2003 survey. J Cataract Refract Surg 2004; 30:892–900
J CATARACT REFRACT SURG - VOL 32, FEBRUARY 2006
235