Delayed toxic anterior segment syndrome after cataract surgery: a case report

Article information

Abstract

Introduction

Toxic anterior segment syndrome (TASS) is an acute, noninfectious postoperative inflammation occurring after anterior segment surgery [1–3]. Previously, this postoperative inflammation that appeared to be noninfectious was often described as sterile endophthalmitis or postoperative uveitis of unknown origin [4,5]. There is still controversy concerning whether these conditions are the same, but TASS has become a major concern for many cataract surgeons due to its sequelae [6,7].

TASS typically develops 12 to 48 hours after surgery and is characterized by corneal edema and accumulation of white cells in the anterior chamber of the eye, with negative Gram stains and negative cultures. Although most cases of TASS can be treated successfully with topical steroids, the severe inflammatory response can cause serious damage to intraocular tissues, such as corneal endothelial loss, trabecular meshwork destruction, and iris damage. Severe anterior segment inflammation with hypopyon or fibrin formation is also commonly observed in TASS, so it is very difficult to differentiate TASS from infectious endophthalmitis. TASS has no infectious signs such as lid swelling, conjunctival injection, eye discharge, or pain. Inflammation involves only the anterior segment within 24 hours after surgery, and it responds well to topical steroids [6,7].

In this report, I introduce the unusual delayed-onset TASS with gray-white opaque deposits on the anterior surface of the intraocular lens (IOL) after uneventful phacoemulsification. In spite of sufficient topical steroid therapy, multiple whitish inflammatory deposits did not disappear. Neodymium:yttrium-aluminum-garnet (Nd:YAG) laser therapy can clear the lumpy deposits completely.

Case Report

A 55-year-old male patient had a 3-year history of primary open angle glaucoma, treated with anti-glaucoma drugs including latanoprost and timolol maleate (Xalacom, Pfizer). He had no past medical history of uveitis or systemic diseases such as autoimmune disease or diabetes.

After the pupil was maximally dilated, lidocaine 2% was instillated for topical anesthesia. Through clear corneal incision, injection of sodium hyaluronate 1.5% into the anterior chamber, standard phacoemulsification and insertion of a foldable hydrophobic acrylic IOL (Sensar, Johnson & Johnson Vision Inc.) in the capsular bag. No drugs were injected in the anterior chamber or subconjunctival space after surgery. At the end of the surgical procedure, an 0.8-mm ribbon of dexamethasone-neomycin-polymyxin B ointment (Maxitrol, Novartis) was applied in the inferior conjunctival sac, and operated eye was appropriately patched with thick gauze and a plastic shield.

On the first postoperative day, the cornea was clear, and mild inflammatory cells (+1) were observed. Moxifloxacin HCl 0.5% (Vigamox, Novartis), prednisolone acetate 1% (Predforte, Allergan) were routinely administered four times daily and bromfenac sodium hydrate 0.1% (Bronuck, Taejoon) was two times.

However, at 1 week postoperatively, anterior chamber inflammatory cells had increased to +4 levels. He did not have lid swelling, conjunctival injection, fibrin formation or pain. Moxifloxacin HCl 0.5% was instilled every two hours and the prednisolone acetate 1% was decreased to once a day in consideration of the risk of infection. To rule out infectious endophthalmitis, Gram stain and 10% KOH mount were performed after aspiration of the aqueous humor, and aspirates were inoculated on blood agar, chocolate agar, and Sabouraud dextrose agar, and in thioglycolate broth. Cultures for aerobes, anaerobes, and fungi were all negative.

At 2 weeks after surgery, he complained of blurry vision (best corrected visual acuity as 20/32) with normal intraocular pressure. Despite the use of frequent topical antibiotics, inflammatory cell did not show any changes, and slitlamp examination showed the multiple, elevated, gray-white opaque deposits on the anterior surface of the IOL (Fig. 1). Under retro-illumination, peculiar greasy oil film coating the anterior optic surface of the IOL in a waved pattern was found (Fig. 1A).

Fig. 1.

Anterior segment photography of patient. (A) A greasy oil film with a wavy pattern (arrow) was seen coating the anterior optic surface of the intraocular lens. (B) Multiple, elevated, gray-white, opaque deposits appeared in a scattered pattern on the anterior surface of the lens at 2 weeks postoperatively. (C) At high magnification, central dense precipitates surrounded by transparent material were definitively observed.

On the presupposition that there would be postoperative toxic inflammation, prednisolone acetate 1% was administered every 1 to 2 hours for 10 days. The anterior chamber reaction decreased, however there was no significant decrease in the number or extent of whitish fluffy deposits.

To remove whitish fluffy deposits, Nd:YAG laser (0.8 mJ) was applied to disrupt only a part of the precipitates on the inferior optic as a trial. The precipitates were very easily removed by laser, and their remnants were dispersed into the anterior chamber. Prednisolone acetate 1% was administered four times a day. Two weeks after laser therapy, no recurrence of deposits was noted at the laser site and the optic surface remained clear (Fig. 2). After complete removal of all deposits with Nd:YAG laser, there was a transient anterior chamber reaction, which was well controlled with 0.1% fluorometholone (Fumeron, Hanlim Pharm) four times a day for 2 weeks. The anterior chamber reaction resolved to nearly normal. There were no adverse events such as increased intraocular pressure and lens pitting. After laser therapy, best corrected visual acuity improved to 20/20 and blurry vision disappeared and there was no recurrence during the 2-year follow-up period.

Fig. 2.

(A, B) Part of multiple deposits (red circle) were very easily disrupted and removed by Nd:YAG laser treatment. There was no recurrence of deposits at the laser sites for 2 weeks, and the optic surface remained very clear. (C, D) Multiple precipitates on the optic surface were successfully removed using a laser, and their remnants were dispersed into the anterior chamber. The surrounding remnants were absorbed completely using a weak topical steroid. Nd:YAG, neodymium:yttrium-aluminum-garnet.

Discussion

The typical hallmark of TASS is a sterile inflammation that starts within 24 hours of cataract surgery, is limited to anterior segment of the eye, and improves with steroid treatment. Although this study revealed unusual clinical findings, we were convinced that the patients had TASS. The inflammation was limited to the anterior segment, and there were more inflammatory cells (+3 to +4) that are not found in uneventful cataract surgery. These inflammatory reactions were not controlled by antibiotics, however were decreased by intensive steroid therapy, even though gray-white material remained on the optic surface. All laboratory tests for infectious etiology were negative.

When the multiple gray-white materials on the acrylic IOL optic were found, late onset endophthalmitis caused by the Gram-positive bacterium, Propionibacterium acnes, fungi, or mycobacteria should be suspected. Specifically, multiple transparent amoeboid deposits with central, elevated, fluffy precipitates gave us a cue concerning bacterial growth on the culture media. To rule out infectious endophthalmitis, aspirates of aqueous humor were stained with special technique and directly inoculated in various culture medium. All results were negative, but we could not confirm these results definitively because majority of P. acnes infection cases are culture-negative [8]. DNA detection by the polymerase chain reaction directed at the 16S rDNA of P. acnes might have been diagnostically useful and might have facilitated proper management [9]. Since long-term bacteria & fungus culture tests for 1-month showed all negative results, I became to believe that it was TASS.

Preoperative central corneal thickness was 562 µm and the endothelial cell count was 2,527 cells/mm². At the time of diagnosis of TASS, corneal edema was not prominent, and best corrected visual acuity was not poor (20/32), thus, corneal thickness was not measured unfortunately. In severe inflammation, corneal edema may be a hallmark of TASS, but in the case of delayed-onset and mild TASS, corneal endothelial cell damage may be insignificant and corneal edema may not appear. Other previous studies indicated that the proportion of corneal edema in late-onset TASS was only 15.6%–19.1% [10,11].

TASS accompanied by elevated gray-white deposits was present at two weeks postoperatively. While most TASS cases are acute and severe, there have been reports of delayed-onset postoperative sterile inflammation. IOSs including the Memory lens [2] and the Verisyse iris-supported phakic IOL [3] have been shown to induce TASS at 1 to 21 days postoperatively. Chemicals used in IOL polishing, cleaning, sterilizing, packaging, or coating have been implicated as causes of toxic inflammation. Another report about late-onset TASS outbreak showed that TASS was identified at postoperative 28.9±19.9 days with anterior chamber cells (92.9%), deposits on the IOL (57.1%), or fibrinous inflammation (35.7%). The cause of TASS was presumed the aluminum and silicon contamination of the IOL surfaces [12]. Recently, two cases of delayed onset TASS after ICL insertion were reported. TASS was detected at postoperative 1 week and exact cause was not identified, just only correlation with naphazoline hydrochloride eye drop [13]. Another report contended that the ingress of ophthalmic ointment after tight patching on a clear corneal wound was responsible for TASS [14]. Ointment-induced TASS was seen at 1 to 5 days postoperatively. Hydrocarbons from petroleum have been thought to cause the inflammation.

After the detection of TASS, I performed a complete evaluation of all medications, fluids, and instruments used during surgery and sterilization protocols in our hospital. There were no protocol violations or problems. However, there was an unusual postoperative finding; a greasy oil film coating the anterior optic surface of the IOL in a wavy pattern. Hence, I assumed that the whitish materials were piles of gray-white inflammatory cells, mingled with the tiny oil immersion from the Maxitrol ointment applied at the end of surgery. These inflammatory cells—such as small round cells, spindle shaped cells, epithelioid cells, and foreign body giant cells—were piled up on the anterior surface of IOL. After topical steroid therapy for 1 month, anterior chamber inflammation was decreased, however, piled up whitish deposits showed no change. Mechanical removal via irrigation/aspiration, or intracameral or subconjunctival steroid injections could be considered. Among various treatment options, I thought that indirect treatment such as topical steroids, intracameral or subconjunctival steroid would not effective due to their lack of penetration to the dense deposits capturing inflammatory cells mixed with the tiny oil immersion. Mechanical removal via irrigation/aspiration was too invasive to performed. Therefore, I chose the Nd:YAG laser. After the dispersion of inflammatory cells into the anterior chamber with Nd:YAG laser therapy, the steroids could resolve the dispersed inflammatory cells effectively. Suspended tiny oil droplets, floating in the anterior chamber after laser treatment, might be absorbed gradually by phagocytosis of trabecular endothelial cells. So, there was no recurrence during the 2-year follow-up period after laser treatment. When performing the laser, the laser beam should be aimed at the front of the IOL to disperse the deposits accumulated on the surface of the IOL. Attention should be paid to various complications such as posterior capsule rupture, IOL pitting, and intraocular pressure spike when performing Nd:YAG laser therapy.

There have been several important studies of intraocular ointment [14–20]. Those reports indicate that the major predictor of clinical severity within the eye is the amount of ointment instilled. Those reports showed that the presence of a small amount of intraocular ointment may not be an indication for removal unless complications directly attributable to its presence are demonstrated. For the ingress of ocular ointment, wound integrity itself is more important than incision method is. There are two reports of postoperative inflammation related with the intraocular ointment after cataract surgery with scleral tunnel incision [15] and clear corneal incision [14]. Werner et al. [14] reported that the use of a clear corneal wound and postoperative topical ointment with tight patching allowed ingress of this material into the anterior segment, causing delayed onset TASS.

I therefore hypothesized that TASS occurred due to a small portion of ointment that accidentally entered the anterior segment, eliciting an atypical cellular reaction. Because there was no need to extract of the IOL for the treatment, I could not determine the composition of the whitish materials with scanning electron microscopy or X-ray microanalsysis [14,15], but I believe the gray-white masses represented piles of greasy inflammatory cells induced by the ingress of tiny oil components. After discontinuation of dressing with ophthalmic ointment postoperatively, no additional cases were found.

This study has several limitations. There was no specific and confirmative analysis about the cause of TASS. And with only one case, the cause of TASS was estimated using circumstantial clues.

I described the unusual delayed onset of mild TASS following uneventful cataract surgery. The pathogenesis of this TASS can be explained as follows; After the ingress of a small amount of ophthalmic ointment, a late foreign body reaction occurred, and gray-white opaque deposits related to the inflammatory cells and ointment capture on the anterior surface of the IOL developed. Nd:YAG laser was effectively used to remove and disrupt the inflammatory cell deposits, mingled with greasy oil material, and thereafter, mild topical steroids controlled the inflammation effectively.

Article information Continued