Toxic retinopathy after phacoemulsification when the cefuroxime dilution is correct: Expect the expected

Abstract

The candor of Zuo et al. is refreshing.1 The authors stated that evanescent macular edema and extensive posterior serous retinal detachment occur with intracameral cefuroxime administered at the conclusion of phacoemulsification cataract surgery. Nevertheless, Zuo et al.’s data do not clarify the size of the referral cohort, the demographic characteristics of the patients, the experience of their surgeons, or the referral pattern constrictions for the 20 study patients. It is unclear at what stage the referrers considered that the visual loss could have been caused by corneal edema, difficult surgery in eyes with high-grade nuclear sclerosis, or preexisting macular edema. The authors acknowledged that the total cefuroxime dose was not standardized because the total volume administered might have varied.1 It is concerning that toxicity can arise from doses similar to or at the authors’ recommendation of 1 mg/0.1 mL, suggesting a narrow therapeutic index. Fortunately, their presumptive diagnosis of toxic retinopathy, generated intraoperatively, appears to be a relatively benign postoperative complication, and the recovery of their study patients is welcomed. It is becoming de rigueur worldwide to address the postoperative bacterial invasion of the anterior chamber with an intracameral injection of dilute cefuroxime, customarily administered at the conclusion of phacoemulsification. However, it is likely, despite published “mega data” to the contrary,2 that these injections are not relevant to addressing the etiopathogenesis of post-cataract endophthalmitis.3,4 Furthermore, morbidity can occur from the injecting cannula itself.5 Even in large studies, several of which have been addressed by our group in the literature,4 the benefit of intracameral intraoperative cephalosporin has not been shown convincingly. There are numerous explanations for this, including that these large studies of endophthalmitis are almost universally retrospective. The surgeon whose patient has had a technically well-performed phacoemulsification procedure and subsequently presents in the early postoperative period with blinding bacterial endophthalmitis might be devastated. This outcome probably does not lend itself to enthusiastic reporting and might minimize the imputed endophthalmitis rate. Moreover, reports of a much higher endophthalmitis rate than expected or “accepted” by analysis of endophthalmitis data can be quite difficult to get into print.4 Our group contends that these studies might not be up to the mark because the question of postoperative inflow of periocular fluids, including tears, mucus, bacteria, debris, topical preserved medications, and general fomites, might not have been addressed. This is pertinent because the literature suggests that endophthalmitis occurs as a result of ingress of infective material from the ocular appendages (in particular the lids, lashes, and lacrimal sac) postoperatively. Furthermore, it is unlikely that the anterior chamber would retain any pathogens given that during routine phacoemulsification the anterior chamber contents are replaced 1400 times by irrigation.4 Today, many surgeons rely on hydrated clear corneal wounds to prevent bacterial ingress. The duration of this effect ranges from 20 minutes to hours postoperatively. In other words, attempted wound impermeability using corneal hydration might be inadequate compared with definitive closure by suture or, for example, by a hydrogel sealant. A prospective endophthalmitis study comparing wound closure with no closure is needed.