A 27-year-old male underwent uncomplicated bilateral LASIK with the Visx Star S4 and IntraLase FS lasers (both from Abbott Medical Optics Inc., Santa Ana, CA) for -2.75 -1.00 X 55 OD and -3.00 D sphere OS (Figure 1). His preoperative BSCVA was 20/20 OU. On the first postoperative day, the patient had a UCVA of 20/20 OU and reported no discomfort, but he had a trace of peripheral inflammation in his left eye. The surgeon increased the steroid from four to eight times daily in this eye, and the patient was scheduled to return for a follow-up visit in 1 week.
On postoperative day 6, the patient presented with a complaint of blurred vision and slight soreness in his left eye. His BSCVA had decreased to 20/70 OS, and the cornea of that eye had dense, focal, anterior haze centrally that was cracked in appearance, was associated with overlying striae, and measured approximately
4 X 4 mm (Figure 2). There also appeared to be thinning associated with this area that was causing a resultant topographical change (Figure 3). The patient's right eye was entirely normal.
The surgeon promptly irrigated the flap interface of the patient's left eye and found no evidence of inflammatory cells. Where it overlay the area of haze, the flap had a jelly-like consistency with poorly adherent epithelium focally.
What is your diagnosis, and how would you proceed?
BRIAN S. BOXER WACHLER, MD
From the history, the patient had diffuse lamellar keratitis (DLK) in his left eye, which prompted the surgeon to increase the steroids to eight times per day. Unfortunately, the patient was not seen on a daily basis to monitor his DLK status. The "smoking gun" of stromal melting from grade 4 DLK is the (1) hyperopic shift, (2) irregular topographic flattening, and (3) striae in the flap giving the appearance of cracks. On postoperative day 6, the "excitement" in the cornea is over; the inflammatory polymorphonuclear cells in the stroma have "gone home." All that is left is the irregular hyperopic astigmatism with corneal thinning and striae.
I think a key message is that DLK should be monitored closely during the first week, because there is a small window of opportunity to perform irrigation before permanent stromal damage occurs.
There is good news. Over the next 3 months, the epithelium will undergo a hyperplastic response that will help fill in the excessive flattening. I would expect the patient's hyperopia to improve along with the irregularity on topography. If there is residual irregular hyperopic astigmatism after 3 months, I would consider two surgical options: wavefront-guided PRK with mitomycin C or hemiconductive keratoplasty (half a circle of conductive keratoplasty spots straddling the axis 160° and placed in the flatter inferotemporal half of the cornea). The latter approach will treat the patient's hyperopia, reduce the amount of astigmatism, and steepen the flat zone on topography.
LOUIS E. PROBST, MD
The patient has experienced a rapid onset of DLK. After performing more than 85,000 LASIK procedures during the past 14 years, I have unfortunately had four cases of this complication. Although this clinical situation has been described as central toxic keratopathy, I believe that it simply represents the most aggressive form of DLK rather than a separate clinical condition. The mild peripheral inflammation visible in the patient's left eye is typical of rapid-onset DLK. The inflammatory cells start on the peripheral edge of the flap and then advance to the central area of the cornea over the next few days unless dosing with topical steroids is increased to every hour at the first sign of inflammation. Even with hourly topical steroids, the DLK can progress in some patients. If there is any sign of DLK's progression on the second or third postoperative days, I have a very low threshold for irrigating the interface, which is usually curative and prevents progression to grades 3 or 4 DLK. Fortunately, aggressive DLK is rare with the IntraLase FS laser: not one of my 15,000 patients undergoing iLASIK (Visx CustomVue [Abbott Medical Optics Inc.] and Intralase) has developed DLK that has required irrigation.
In this case, the DLK started on the first postoperative day and then likely progressed to grade 4 DLK by the fourth or fifth postoperative day. All of my cases of grade 4 DLK progressed at this aggressive rate. When this patient returned for his postoperative visit on day 6, the inflammation had subsided, so few cells were in the interface, which left the central flap striae and scarring. The central thinning is also typical of grade 4 DLK. When lifting the flap at this time, I have found its center to be very soft and fragile and the epithelium to be quite loose.
At this point, the management is a long-term strategy. The opportunity to irrigate the interface in order to eliminate the inflammation when the DLK was only grade 2 has been lost. Now, there are no quick fixes to resolve the patient's problems, although, fortunately, successful treatment is still possible.
The first and most important step is for the surgeon to be fully engaged in the patient's care. I would make it clear that I will do whatever is necessary and will see the patient as many times as needed to resolve the problem. I would tactfully inform him that full resolution could take over a year. He should be seen weekly for at least 1 month and monthly for at least 1 year.
I would taper the topical steroids over several weeks. Although no inflammation was evident on postoperative day 6, the steroids will cover any residual subclinical inflammation and reduce the progression of the central corneal scarring. I would monitor the IOP to make sure the patient is not a steroid responder. By the second postoperative week, the vision in this eye will start to become hyperopic with astigmatism from the central striae. At this point, I would fit the eye with a tight, soft hyperopic contact lens to correct the refraction, improve the vision, and induce some corneal steepening, which can limit the progression of the hyperopia. I would then allow the cornea to stabilize for about 6 months. When the refraction stabilized, I would perform a customized hyperopic PRK with mitomycin C for the residual refraction. This procedure could be repeated in another 6 months if necessary.
The central corneal haze and striae will gradually fade by 12 months postoperatively. With the customized PRK treatment of the residual refraction, the patient's UCVA can usually be corrected to close to 20/20 in my experience.
STEPHEN COLEMAN, MD
This case highlights that even trace amounts of very peripheral inflammation on the first postoperative day should not be taken lightly. These eyes require diligent follow-up and attention, regardless of the way in which the initial flap was created.
Having this particular patient return for follow-up on the second and/or third postoperative days would have greatly facilitated making the correct diagnosis to determine the proper course of action. As it stands, the most likely diagnosis is central toxic keratopathy, otherwise known as central flap necrosis.1 DLK is a distant second for the diagnosis, and the approaches to treatment differ somewhat. The key to the diagnosis is the critical finding of a jelly-like consistency to the epithelium and anterior flap.
Oddly, some surgeons believe that central toxic
keratopathy—first described and most accurately named by Robert Maloney, MD2 (also oral communication, May 2005)—is a missed or late diagnosis of DLK. In my experience, the two complications behave quite differently and have unique characteristics. Based on the six cases I have observed (three eyes in my practice and three eyes seen on referral), what starts out looking like a rapidly accelerating case of DLK turns out to be something quite different and relatively predictable.
For instance, early, aggressive irrigation of the interface in a case of true central toxic keratopathy reveals that there are no inflammatory cells to irrigate, and the slit-lamp examination on the following day is effectively unchanged. The central striae and opacification can be quite dramatic. In addition, at the time of irrigation, careful attention to the anterior aspect of the cornea will show its boggy, jelly-like consistency with poorly adherent epithelium that is not caused by prior drops. This finding can also be demonstrated at the slit lamp with proparacaine and a Weck-Cel sponge (Medtronic ENT, Jacksonville, FL). The result of the interface's irrigation is the key to making the diagnosis (ie, no cells).
Over the course of several weeks to months, the slit-lamp examination will show a hazy anterior flap that improves slowly, and the refraction, which typically starts out hyperopic, will likely remain so. Steroids are not indicated. The patient's UCVA will consistently be diminished by up to four lines or more in the short term, but an aspheric hyperopic soft contact lens can help during this period.
All three of the eyes that I have managed from start to finish required a hyperopic retreatment, and they all currently have UCVAs of 20/25 or better. Confidence in the diagnosis and a review of the two articles I have cited1,2 can greatly assist in the management of these patients.
Section editor Karl G. Stonecipher, MD, is the director of refractive surgery at TLC in Greensboro, North Carolina. Parag A. Majmudar, MD, is an associate professor, Cornea Service, Rush University Medical Center, Chicago Cornea Consultants, Ltd.
Stephen Coleman, MD, is the director of Coleman Vision in Albuquerque, New Mexico. He acknowledged no financial interest in the products or companies mentioned herein. Dr. Coleman may be reached at (505) 821-8880; email@example.com.
Brian S. Boxer Wachler, MD, is the director of the Boxer Wachler Vision Institute in Beverly Hills, California. He acknowledged no financial interest in the products or companies mentioned herein. Dr. Boxer Wachler may be reached at (310) 860-1900; firstname.lastname@example.org.
Louis E. Probst, MD, is the national medical director of TLC The Laser Eye Centers in Chicago; Madison, Wisconsin; and Greenville, South Carolina. He is a consultant to Abbott Medical Optics Inc. and TLCVision. Dr. Probst may be reached at email@example.com.
- Hainline BC, Price MO, Choi DM, Price FW Jr. Central flap necrosis after LASIK with microkeratome and femtosecond laser created flaps. J Refract Surg. 2007;23(3):233-242.
- Sonmez B, Maloney RK. Central toxic keratopathy: description of a syndrome in laser refractive surgery [published online ahead of print December 19, 2006]. Am J Ophthalmol. 2007;143(3):420-427.