Transient Corneal Edema in the Second Eye during Immediately Sequential Bilateral Cataract Surgery under General Anesthesia: Two Case Reports

Dear Editor,

Immediately sequential bilateral cataract surgery (ISBCS) offers advantages including convenience, reduced recovery time, and lower costs. It is crucial to assess the patient’s ability to cooperate and their propensity for involuntary movements; in severe cases, general anesthesia should be considered. Written informed consent for publication of the research details and clinical images was obtained from the patient.

Corneal complications, particularly transient corneal edema (TCE), are observed sequelae following cataract surgery and other intraocular procedures, potentially compromising surgical outcomes and visual recovery. TCE is typically presented as reversible corneal clouding or Descemet membrane folds, most notably observed in the early postoperative period. Risk factors include preexisting endothelial dysfunction, intraoperative endothelial trauma (e.g., Descemet membrane detachment), transient postoperative intraocular pressure elevations, and prolonged surgical time increasing endothelial stress [1].

However, the occurrence of TCE in the second eye during ISBCS under general anesthesia before the initiation of surgery has rarely been reported. This report aims to investigate the potential causes underlying this phenomenon to improve understanding and management of such rare complications during ISBCS under general anesthesia.

Case 1 is a 62-year-old female patient with no history of ocular or systemic disease who presented with bilateral blurred vision. Her corrected distance visual acuity (CDVA) was 20/125 and 20/63 in the right and left eye, respectively. Preoperative examination showed no corneal edema, and the endothelial cell density (ECD) was within the normal range. Following the induction of general anesthesia due to claustrophobia, a preoperative antisepsis protocol was implemented, using 2% chlorhexidine gluconate in 70% isopropanol (Hexitanol, Firson) diluted 1:20 in saline, followed by 5% povidone-iodine for eyelid margins and ocular surface, 10% povidone-iodine for the skin, and finally 83% alcohol applied to surrounding skin except the periorbital area. This approach was based on prior evidence supporting the potent antimicrobial effects of both alcohol-based skin preparations and chlorhexidine, which are widely regarded as effective for reducing surgical site infections [2]. While 0.1% chlorhexidine is a well-tolerated antiseptic for ophthalmic procedures [3], recent studies have highlighted that when chlorhexidine and alcohol-based antiseptics inadvertently contact the ocular surface, they are associated with ocular toxicity including corneal damage and epithelial disruption particularly under general anesthesia where protective blinking is absent [4]. To minimize this risk, antisepsis was not applied simultaneously to both eyes. After uneventful surgery and patching of the right eye, the left eye was disinfected and draped separately using a new set of sterile instruments and supplies, following the same antiseptic protocol as used for the first eye.

Central corneal haziness was then noted, which worsened during continuous curvilinear capsulorhexis (CCC) but gradually improved during cortex removal (Fig. 1A), allowing for successful intraocular lens (IOL) implantation. To manage intraoperative corneal haze, dispersive ophthalmic viscosurgical device (OVD) was applied to protect the endothelium. Phacoemulsification was performed with reduced ultrasound power. Postoperatively, CDVA improved to 20/20 in both eyes.

Fig. 1.

Intraoperative corneal haziness and animal model of antiseptic-induced corneal edema. (A) Case 1 and (B) case 2. The top row shows intraoperative views during nucleus removal in the right and left eyes, respectively. Central corneal haziness is markedly more evident in the left eye compared to the right eye. The bottom row shows post–intraocular lens implantation views of the right and left eyes, respectively. The previously noted corneal haziness in the left eye has resolved. (C) Central corneal thickness measured at 5, 10, 20, and 30 minutes after instillation of 50 μL of 5% povidone-iodine, 50 μL of 0.1% chlorhexidine gluconate, 50 μL of normal saline, and a smaller volume of 7 μL of 83% alcohol in four eyes (two rabbits), consecutively; each eye was instilled a different solution (50 μL, about the volume of one drop). After 30 minutes, the cornea thickened with alcohol administration but no significant difference compared with the baseline corneal thickness was found in case of the other antiseptics. (D) Central corneal thickness with 2 μL of 83% alcohol administration was monitored for up to 2 hours. The cornea thickened for the first hour and then began to decrease shortly after. (E) The photographs show the corneal condition at each time point after instillation of 83% alcohol. Immediately prior to taking the photographs, normal saline was instilled to prevent dry eye.

Case 2, performed 6 months after the previous surgery, is a 61-year-old female patient with similar findings who also underwent ISBCS under general anesthesia. As in case 1, the right eye was operated first, and after left eye disinfection and draping, central corneal haziness was observed. The haze worsened during CCC and improved during cortex removal (Fig. 1B), allowing safe IOL implantation. Postoperative recovery was uneventful; CDVA was 20/20 in both eyes.

In case 1, specular microscopy and pachymetry parameters remained stable at 3 months postoperatively, indicating no lasting endothelial compromise. In case 2, uncorrected visual acuity was 20/20 with a clear cornea on slitlamp examination at 1 month, but the patient was lost to follow-up thereafter (Supplementary Table 1).

We present two cases of TCE that occurred in the second eye during ISBCS under general anesthesia. No intraoperative trauma, OVD instability, or anesthesia-related toxicity was evident. Both patients had normal ECD. We hypothesized alcohol-based antiseptics inadvertently contacted the contralateral eye during draping. The antiseptic regimen included diluted chlorhexidine and 83% alcohol. Previous studies support corneal toxicity from chlorhexidine and alcohol under general anesthesia [4]. Although classic signs like epithelial disruption or hyperemia were absent, mild toxicity from residual alcohol likely explains the transient haziness. Similar corneal toxicity under general anesthesia has also been reported in nonocular surgeries using alcohol-based antiseptics [5]. Irrigation during cortical removal may have diluted residual antiseptics and resolved the haziness.

To explore the cause of intraoperative corneal edema, we conducted an animal study using two New Zealand white rabbits. Each eye received either 5% povidone-iodine, diluted 0.1% chlorhexidine gluconate, saline, or 83% alcohol. Central corneal thickness was measured serially. Only alcohol induced a transient increase in thickness without epithelial damage (Fig. 1C). One month later, a small amount of 2 μL of 83% alcohol was instilled in the eye previously instilled with 5% povidone-iodine separately, and changes to the corneal thickness were observed. A reduction in corneal edema was noted around the 1-hour mark. This quantity of alcohol did not result in any evident epithelial damage or hyperemia in the rabbits (Fig. 1D, 1E). All animal experiments adhered to institutional guidelines and were approved by the Inha University Institutional Animal Care and Use Committee (No. INHAIACUC 220816-833).

While chlorhexidine and alcohol-based agents are effective for skin antisepsis, their accidental exposure to the ocular surface may result in corneal epithelial toxicity, particularly under general anesthesia due to the absence of protective blinking [3]. In contrast, 5% povidone-iodine applied to the conjunctival sac for at least 3 minutes without irrigation has demonstrated broad-spectrum antimicrobial efficacy and a favorable safety profile. This approach is currently regarded as the most effective prophylactic measure against postoperative endophthalmitis in cataract surgery. Following this, we revised our antiseptic protocol to exclude alcohol-based agents and fully occlude the contralateral eye during preparation. No further TCE cases were observed after implementing this change, supporting the hypothesis that even minimal ocular exposure to alcohol-based antiseptics may induce transient endothelial dysfunction under general anesthesia.