Answers are grounded in this paper and cite their sources.
The relationship between unsutured clear corneal tunnel incisions and an increased incidence of infection after cataract surgery remains uncertain. Although a large study from the Bascom Palmer Eye Institute reports no greater incidence of endophthalmitis with corneal incisions than with scleral incisions,1 recent literature presents evidence that postcataract endophthalmitis is more likely with corneal incisions.2–4 Indeed, laboratory models indicate that corneal tunnel incisions do not provide hermetic sealing under certain conditions.5–7 These investigations suggest that the incisions may be competent at physiologic levels of intraocular pressure (IOP) but fail when IOP is lowered. Critics of the reports, in which human cadaver eyes are used, mention that postmortem eyes lack the endothelial pump mechanism thought to be partly responsible for maintaining incisional self-sealing. Nonetheless, there is concern regarding the relationship between clear corneal incisions and rates of postoperative infection, which suggests that we need to carefully evaluate the potential. The article by Wallin et al. in this issue (pages 735–741) reveals that all cases of endophthalmitis at the Moran Eye Center at the University of Utah from 1996 through 2002 were associated with unsutured clear corneal incisions. During this period, in accord with the record review, no cases of infection were associated with sclerocorneal incisions. What factors might account for a greater risk for infection after corneal incisions? One might consider the absence of an overlying, separate external conjunctival closure as potentially causal. However, this would imply that sclerocorneal incisions routinely have a meticulously and hermetically sealed conjunctival closure overlying the external aspect of the scleral tunnel; this is a seemingly unlikely scenario. Moreover, in virtually all anterior segment procedures, a clear corneal paracentesis is performed and there has generally been little to no association between well-constructed, small side-port incisions and increasing rates of postoperative infection. If paracenteses are at low risk for infection and clear corneal cataract incisions potentially present a higher risk for endophthalmitis, what factor(s) could account for this paradox? Wound architecture is the logical answer. One theory suggests that ocular hypotony shortly after surgery allows corneal incisions to be deformed easily, inducing wound leak with further hypotony and a resultant pressure gradient from the outside in. This suggested pathway provides a portal for bacteria to contaminate the anterior chamber. If deformation susceptibility is the culprit, appropriate wound construction should be the antidote. Work by Ernest and coauthors8 has convincingly demonstrated that cataract tunnel incisions that are square or nearly square in surface architecture are significantly more resistant to external deformation than those that are rectangular. In addition, the Langerman-style, deeply grooved preincision has been shown to resist deformation.9 Nonetheless, no literature references support reduced rates of infection associated with these principles. Construction of corneoscleral tunnel incisions generally includes careful and controlled dissection of the tunnel following a specific and often marked template (Figure 1). After tunnel dissection, the chamber is entered in clear cornea, creating the self-sealing valve (Figure 2). In this manner, the surgeon has the opportunity to construct an incision with predictable architecture. Clear corneal tunnel (and “near clear”) incisions, by contrast, are fashioned with sharp (often gem) blades that create the tunnel and the chamber entry in 1 motion. As a result, the surgeon tends to have less control over the configuration of the incision. This, added to the fact that corneal incisions originate more centrad than sclerocorneal incisions, contributes to rectangular rather than square surface architecture (Figure 3).Figure 1.: Dissection of corneoscleral tunnel incisions generally use lamellar dissection with steel blades, conforming to a specifically measured template.Figure 2.: Following tunnel dissection, a sharp keratome is used to perforate Descemet's membrane and enter the anterior chamber. Typically, the surface architecture of the incision is almost square, representing equal width and length of the tunnel.Figure 3.: In this clear corneal cataract incision, the incision is rectangular in its architecture; the width of the tunnel is roughly 2 times that of the tunnel length. As a result, 2 radially oriented sutures were placed at the time of surgery.Several factors determine the length of the corneal tunnel; these include sharpness of the blade, IOP, thickness of the corneal tissue, and angle of approach. Care must be taken by the surgeon to ensure that corneal tunnel incisions are constructed to match the concept of square surface architecture to exhibit adequate resistance to deformation. Another concern is (self) sealing of incisions. The cornea is less tolerant of stretching and distortion than the sclera. If the cataract incision is inadequate in width and subjected to stretching by surgical instrumentation, it is less likely to maintain its integrity. In my view, an entry wound that is too narrow and distorted during surgery will be more apt to leak than a modestly larger but less traumatized incision. This may also be true for bimanual microincision cataract surgery, as suggested by Buratto and Giardini in the August 2004 Consultation Section of JCRS.10 I, too, remain concerned that the unsleeved, rigid round tubes used for bimanual microphaco (irrigating choppers and phacoemulsification needles) distort the small slit incisions used for bimanual cataract surgery, increasing chances of postoperative leak. It would seem, therefore, that incision design and construction play pivotal roles in increased rates of infection with corneal tunnels. Poorly constructed and distorted wounds could contribute to a greater chance of postoperative anterior chamber contamination. Unfortunately, details of or standards for incision configuration were not offered in the Wallin study. For the future, one might consider methods to standardize the clear corneal cataract incision to create architecturally consistent and truly self-sealing incisions. Femtosecond lasers may be helpful and are presently under investigation for this purpose. Biologic tissue adhesives may play a future role. In the interim, surgeons who use corneal tunnels must pay particular attention to incision details and ensure proper sealing at the close of surgery. Any incision suspected of incompetence (including side-port paracenteses) should be considered for suturing, bandage lenses, etc., as the Wallin et al. investigation strongly emphasizes the relationship between incision leak and infection potential. Presently, surgeons often use stromal hydration to enhance wound sealing, although the efficacy has not been fully established. Techniques for this vary, but it is essential that the roof as well as the sides of the incision be subjected to hydration. Although I have used a curved cannula for this purpose for many years, Wong (Michael Wong, MD, ASCRS Cataract Eye Mail, 2004) has developed a method of hydrating the incisional tissue via external tracts. At the close of surgery, my practice is to establish IOP at a physiologic level and perform Seidel testing on all incisions to establish proof of closure. With this strategy, I have not noted a case of hypotony or leakage after clear cornea cataract surgery over a 12-year period. Could factors other than wound architecture play a role in infection rates? Could either incisional method be responsible for a greater chance of intraoperative anterior chamber contamination? That the patient's periocular bacterial flora is the source of microbes in the usual sporadic case of bacterial endophthalmitis has been well established.11 Careful draping of the eyelid margins in combination with chemoprophylactic antisepsis is likely to reduce the chance of anterior chamber contamination during surgery. However, in the Wallin et al. study, use of an eye patch was associated with a reduced risk for postoperative infection. Why would an eye patch reduce rates of infection? Although the data analysis did not test for a relationship between clear corneal incision, topical anesthesia, and absence of patching, one might speculate that it is more difficult to adequately drape the lashes and lid margins when topical anesthesia is used. Given that the eyelids are more difficult to control without orbicularis akinesia, draping efficacy may play a role in infection rates. Hence, having patients under injection anesthesia, with flaccid lids, might allow a better draping technique and a reduced likelihood of intraoperative contamination. As a rule, patients who receive injection anesthesia require an eye patch, while those under topical anesthesia are typically not patched. It would be interesting to evaluate the data in the Wallin et al. study with regard to rates of infection when form of anesthesia, use of patching, and type of incision are analyzed together. Topical anesthesia might be an indirect cause of a greater risk for postoperative endophthalmitis, unless the surgeon develops a draping method specific and efficacious for that anesthetic method. The study by Wallin et al. calls our attention to the need for rigid control of surgical asepsis, particularly with respect to incision management. It is interesting to note that some surgeons report no greater rate of infection using clear corneal incisions,10 while others, including the Wallin investigators, suggest the opposite. In my view, appropriate use of aseptic methods including careful draping, chemoprophylaxis, and, particularly, clear corneal incision design, construction, and sealing should minimize risks for infection.
No takes yet. Share an insight, caveat, or question.
Samuel Masket (2005) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: