Link between endophthalmitis, clear corneal incisions debated

One surgeon said there is a greater risk of endophthalmitis with clear corneal incisions, but another surgeon noted that there may be other reasons for the complication increase.

Peter J. McDonnell

The incidence of post-cataract endophthalmitis has been increasing in the United States, but there is disagreement regarding the role clear corneal incision surgery may play.

“We believe that it may be possible in some eyes for surface fluid to travel along these clear corneal incisions to gain ingress to the anterior chamber prior to corneal wound healing and closure of the epithelial defect,” said Peter J. McDonnell, MD, director of the Wilmer Eye Institute.

“There is no question that the introduction of clear corneal incision surgery was a significantly positive contribution in terms of ease, comfort and speed of recovery, and visual rehabilitation for our patients,” Dr. McDonnell said. “However, despite the increased efficiency of surgery and reduced operating times, we still observe blinding complications like endophthalmitis. Some have even suggested that endophthalmitis is on the increase.”

Gradual up-slope

To determine why the number of endophthalmitis cases is on the rise, Dr. McDonnell and colleagues performed a systematic review of published, peer-reviewed journals. They also analyzed the Medicare claims database.

After fluorescein staining, the seal of the clear corneal incision is tested by depressing the posterior lip of the incision with a finger. Image: Fine IH

The study’s biostatistician concluded that “there was a down-slope from 1963 to 1991,” Dr. McDonnell said during a symposium on endophthalmitis at the American Academy of Ophthalmology meeting. “During that period, there was a gradual decline in the incidence of acute endophthalmitis. But since 1992, there has been a gradual up-slope. Within individual years, though, there have been significant variations in the reported incidence.”

In the reports that describe the type of incision created, “there was a substantially greater risk of endophthalmitis – slightly more than two cases per 1,000 – when using clear corneal incision surgery compared to scleral tunnel or limbal incisions,” Dr. McDonnell said.

In a small study from St. Louis, patients who underwent cataract surgery with a corneal incision were 3.4 times more likely to develop endophthalmitis than patients with scleral tunnel incisions. Similarly, a recent multicenter, randomized study found that the relative risk of endophthalmitis was 4.6 times higher in patients who underwent corneal incisions compared to sclerocorneal incisions.

“While these and other studies provide evidence for an association between incision type and endophthalmitis, most are based on observations from a single center or involve only a few cases. Hence, they cannot be generalized to an entire population,” Dr. McDonnell said.

Medicare claims data was used to identify cataract surgery cases from 1994 through 2001 based on primary procedure codes for both inpatient and outpatient procedures. For this 8-year period, the endophthalmitis rate was 2.15 cases per 1,000. “From 1994 to 1997, the rate was approximately 1.8 cases per 1,000. But during 1998 to 2001, that rate increased to about 2.5 cases per 1,000,” Dr. McDonnell said. “This represents a statistically significant increase.”

Ingress of bacteria

Clear corneal incisions “may permit ingress of bacteria during the postoperative period,” Dr. McDonnell said. To demonstrate this, he and his associates used the Miyake technique to observe the creation of clear corneal incisions in eye bank eyes. “We removed the iris and uveal tissue to allow us to observe the internal aspect of the wound as clearly as possible,” Dr. McDonnell said. “We then applied very dark India ink particles to the surface of the cornea.” The investigators also applied digital pressure to simulate a patient instilling eye drops.

Videomicrographs of the eye bank eyes showed tiny jets of India ink entering the anterior chamber in response to removing digital pressure from the globe. “Histology indicates that the India ink was able to track in along the incisions and become trapped in some of the severed edges of the stromal lamellae,” Dr. McDonnell said.

The investigators also videotaped cataract surgery in living eyes. “In some eyes there was some bleeding from the limbal capillary bed,” Dr. McDonnell said. “Essentially, we used this blood as a dye to mark the precorneal tear fluid. We then pressed on the globe to detect any evidence of ingress of blood-stained aqueous into the eye. We showed that a jet of blood from superficial limbal vessels can travel through the clear corneal incision and be visually detected in the anterior chamber as digital pressure is applied to the globe and released.”

Dr. McDonnell said he thinks that more attention must be directed toward a better understanding of the dynamics and mechanisms of corneal incisions, especially in the early postoperative period.

Other factors influence risk

“There is no question that there is an increased incidence of infectious endophthalmitis following cataract surgery,” said I. Howard Fine, MD, a clinical professor of ophthalmology at the Casey Eye Institute during the AAO symposium. “We also know there is an increased utilization of clear corneal tunnels, for which there is a learning curve.”

Changing resistance to antibiotics is another component of increased risk of endophthalmitis, he said. “Overwhelmingly, gram-positive organisms are associated with postoperative endophthalmitis,” Dr. Fine said. “There is a need for the fourth generation of fluoroquinolones.”

Despite the increased use of clear corneal incisions, two published studies from Sweden document the lowest levels ever of postoperative endophthalmitis in that country, he said. “This is due to the use of prophylactic intracameral cefuroxime,” Dr. Fine said. “In Sweden, they are experiencing the exact opposite trend of what is happening in the United States.”

Dr. Fine’s group practice has also bucked the trend. “We have gone at least 8 years and over 7,000 cases without a single case of endophthalmitis,” Dr. Fine said. “We don’t think this is good luck. We believe it is because of our concerted attention to detail. Other surgeons have been recently quoted in the literature as having no increased incidence of endophthalmitis, with a continued preference for clear corneal incisions. Clearly, not all clear corneal incisions are the same.”

Study concerns

Dr. Fine said he had two reservations about Dr. McDonnell’s studies. “We have mechanical factors that promote sealing with valve-like architecture, and we have physiologic factors with corneal endothelial pumping,” Dr. Fine said.

“One has to question the status of the endothelial pump used in the cadaver eyes. We also think pinpoint pressure is highly nonphysiologic,” he said. “One published report found that 20% of eyes measured 2 hours postop had IOP below 10 mm Hg. I thought that data was spurious, so I looked at our last 100 cases for which we had measured IOP within 2 hours of surgery. I discovered that 15% of those patients had IOP lower than 10 mm Hg. But neither the published report nor our own experience showed an increase in endophthalmitis.”

Recent emphasis on speed in cataract surgery “can result in compromised technique, unless there is attention to detail,” Dr. Fine said. In preparing the surgical field, he said, “We use 5% povidone iodine, which has been documented to reduce the flora in the surgical field. We also always evert the lashes with steristrips, so they are lying against the skin.” In addition, Dr. Fine’s practice drapes over the meibomian orifices and uses a wick in the lateral canthus to drain fluid away.

“We believe the temporal corneal periphery is the best incision location, in part because it neutralizes the force of lid blink,” Dr. Fine said. “It is almost impossible to distort the incision with lid blink and squeezing of the lid temporally.” Dr. Fine’s practice also favors use of limbal relaxing incisions as opposed to operating on the steep axis for astigmatism control.

Viscoelastic and cannula

“We think that the use of viscoelastics is important in incision construction,” Dr. Fine said. “If you replace aqueous as you inject viscoelastic, you have a stable, firm eye, which allows for reproducible incision construction and proper architecture. We also prefer a single plane incision because it provides a better valve and maximizes endothelial pumping. We don’t like grooving.” Dr. Fine’s practice employs trapezoidal blades and trapezoidal incisions to create incisions at least 2 mm long and 2.5 to 3.5 mm wide.

“We feel you should never grasp the superior lip of the incision with forceps because you can abrade the epithelium. Once that occurs, your fluid barrier is lost,” Dr. Fine said.

Instead, he said he uses a cannula to instrument the anterior chamber. “We like beveled tips, introduced bevel down because we don’t need to instrument the superior lip,” Dr. Fine said. “We also think power modulations are very important because they allow us not to compromise the incision with thermal insult.”

Incision closure should always be accompanied by stromal hydration without overpressurizing the eye, according to Dr. Fine. “If we have abraded the epithelium over the incision, we use a soft contact lens. We suture if there is ever a question,” he said.

Dr. Fine also tests every incision with fluorescein dye. “It is a much more physiologic test than pushing with a pinpoint,” he said.

Dr. Fine said he is a strong proponent of antibiotics. Preoperatively, he recommended a fourth-generation fluoroquinolone, at least four times a day for 3 days. “Then, on the day of surgery, perhaps every 2 hours. Some data indicates that preloading is important,” he said. Postoperatively, Dr. Fine said he prescribes antibiotics four times a day. “Some people think the first day should be every 2 hours,” he said.

Despite his enthusiasm for clear corneal incision, “I think you should use the incision you can perform with the most reproducibly safe results,” Dr. Fine said.

For Your Information:

• I. Howard Fine, MD, is a clinical professor of ophthalmology at the Casey Eye Institute at Oregon Health & Science University in Portland and in clinical practice with Drs. Fine, Hoffman & Packer, LLC, at 1550 Oak St., Suite 5, Eugene, OR, 97401; 541-687-2110; fax: 541-484-3883; e-mail: hfine@finemd.com; Web site: www.finemd.com.
• Peter J. McDonnell, MD, is director of the Wilmer Eye Institute at Johns Hopkins University of Medicine. He can be reached at Wilmer Ophthalmological Institute, 600 N. Wolfe St., Baltimore MD 21287-0005; 443-287-1511; fax: 443-287-1514; e-mail: pmcdonn1@jhmi.edu; Web site: www.wilmer.jhu.edu.

• Bob Kronemyer is an OSN Correspondent based in Elkhart, Ind.