All's Well That Ends Well: Managing Unexpected Complications

During my career as an ophthalmologist, I have seen my share of complicated cases. In this article, I describe three recent cases in which I had to manage an unexpected complication.

CASE PRESENTATION NO. 1
A 52-year-old woman requested LASIK for the correction of a myopic defect affecting both eyes. The visual acuity in her right eye was 20/20 (refraction, -4.50 -1.00 X 180), and the keratometry reading was 40.50 X 180° and 42.00 X 90°. The PMetrics pupillometer (Ivis Technology, Taranto, Italy) showed that her pupils were 5.33 mm in scotopic conditions and central ultrasound pachymetry was 655 µm. Schirmer's test results were within normal values, and the patient did not complain of problems with lacrimation. The patient's intraocular pressure (IOP) was also within normal limits. Aberrometry showed no aberrations (root mean square, <0.4 OU).

We agreed to perform laser vision correction for the patient's myopic defect. She was told that the intended correction would affect only her distance vision and that her presbyopia would remain unchanged. After the patient agreed, I scheduled her for LASIK using the Technolas 217 Z-100 excimer laser (Bausch & Lomb, Rochester, New York) with a planoscan algorithm. The Hansatome Zero Compression microkeratome (Bausch & Lomb) would be used with a 160-µm plate and a 9.5-mm ring.

On the day of surgery, I treated her left eye first; this operation was straightforward, and the defect was treated according to plan.

HOW WOULD YOU PROCEED?
Proceedings for the right eye were not as smooth. After making sure the ring of the microkeratome was positioned on the eye, suction was activated and optimal, and the globe was well exposed, I commenced cut. Approximately one-quarter into the cut, the patient moved, and suction to the ring was lost, interrupting the cut. I examined the portion of the cut that was created—a half-moon shape of approximately 2 to 3 mm in the lower part of the cornea, not reaching the pupil.

I decided that because the cut had not reached the pupil's edge, I could proceed immediately with another cut. This time, I used the 30 KHz IntraLase femtosecond laser (Advanced Medical Optics, Inc., Santa Ana, California), the most advanced vsrsion of the machine available at the time of the procedure. My goal was to continue the cut from the original section but at a reduced depth (120 µm) to avoid the previous cut. The diameter of the cut was 9 mm.

I created a 260-µm superior pouch to allow the gas bubbles to escape and then centered and positioned the laser's suction ring. Good suction was noted, and the cone was lowered into position and the up-down cut performed in the 12- to 6-o'clock direction. The cut proceeded as planned for approximately four-fifths of the desired length. When it reached the previous cut, something strange occurred: The gas bubbles were normal in the uncut portions of the stroma, but the laser appeared to be cutting air, possibly because the cornea had already been cut.

If this were your patient, how would you proceed?

HOW I PROCEEDED
Instead of raising the flap, I postponed the operation for 1 week. At that time, I then made another cut with the IntraLase femtosecond laser at 180 µm and created a 100 X 260 µm pouch. (In hindsight, this was a bad decision; I should have raised the flap and continued with the ablation in the first operative session.)

This time, everything appeared to have gone according to plan; however, another surprise arrived when the flap was raised. I used the two-spatula technique to raise the flap from up to down. It became extremely difficult to raise the flap when I reached the hot spot (ie, where the three cuts met). The new cut intersected the interface of one of the previous cuts—most likely the cut performed with the microkeratome. After raising the flap, I proceeded to treat the patient with the same planoscan algorithm that I used in her left eye, as the optical zone was still intact.

FOLLOW-UP
On postoperative day 1 examination, the patient's UCVA was 20/25. Two weeks later, her UCVA was 20/20 and BCVA was 20/16 with 0.50 diopters of sphere (DS). The edges of the flap were slightly stressed at the 5- and 7-o'clock positions. Three months later, the patient's UCVA was 20/16.

CASE PRESENTATION NO. 2
A 37-year-old woman requested LASIK to correct a severe myopic error that affected both eyes similarly. Visual acuity in the patient's right eye was 20/20 (refraction, -10.00 -2.00 X 175), and the keratometry reading was 41.00 D axis 175°/43.00 D axis 85°. Examination with the PMetrics pupillometer was 5.71 mm in scotopic conditions, and the central ultrasound pachymetry was 494 µm. Schirmer's test was within normal values. The patient did not complain of any problems with lacrimation. Her IOPs were also within normal limits.

The patient's white-to-white distance (Orbscan; Bausch & Lomb) was 11.80 mm, and endothelial cell count was 3,000 cells/mm2. The Visante OCT (Carl Zeiss Meditec AG, Jena, Germany) measured the anterior chamber depth as 3.36 mm without the corneal thickness; the iridocorneal angle was approximately 30°.

I decided to implant a toric phakic IOL (Visian Toric ICL; STAAR Surgical, Monrovia, California). The lens had the following parameters, as suggested by the manufacturer's nomogram: refraction, +15.50 +2.50 X 94; optic zone diameter, 4.6 to 5.5 mm; overall length, 12.5 mm.

Surgery was performed under topical anesthesia, and the operation proceeded according to plan with no problems arising. The lens was positioned on the planned axis.

The day after surgery, the patient's UCVA was 20/40 and 20/20 with pinhole. Her IOP was 14 mm Hg. The IOL was slightly decentered inferonasally.

HOW WOULD YOU PROCEED?
One week postoperatively, I had a refractive surprise. Refraction revealed 3.75 D of astigmatism. I dilated and examined the eye, and the lens appeared to have rotated 90° from its intraoperative position. I rotated the lens and repositioned it correctly; however, 1 week later, the lens again appeared to have shifted 90° from its calculated position.

If this were your patient, how would you proceed?

HOW I PROCEEDED
At this point, I decided to replace the lens with a -11.00 D Artisan IOL (Ophtec BV, Groningen, Netherlands) and correct the patient's astigmatism with limbal relaxing incisions. The lens was implanted under sub-Tenon's anesthesia.

Seven months after the replacement of the IOL, the patient's UCVA was 20/25 and 20/20 with a refraction of 0.00 +1.50 X 150. The endothelium was not damaged, with a stable cell count at 3,000 cells/mm2. The crystalline lens is transparent.

In all probability, the lens rotated because it was slightly too short for the patient's eye. Unfortunately, at this time, there is no reliable method for measuring the sulcus-to-sulcus distance. Therefore it is preferable, under the appropriate preoperative conditions, to implant a one-size-fits-all IOL.

CASE PRESENTATION NO. 3
A 72-year-old man was scheduled for cataract surgery in his right eye. Approximately 25 to 30 years earlier, he had undergone radial keratotomy (RK). He had also undergone peripheral argon laser barrage in both eyes. There were 12 radial incisions in his right eye, the one requiring cataract surgery.

The patient's IOP was 21 mm Hg with air-puff tonometry and 19 mm Hg with contact tonometry. Colvard pupillometry was 5 mm, visual acuity was 20/40 with the refraction of -4.50 DS, and endothelial cell count was approximately 1,000 cells/mm2. We observed mild diffuse myopic chorioretinosis.

After straightforward cataract extraction performed under topical anesthesia, a one-piece AcrySof IOL (Alcon Laboratories, Inc., Fort Worth, Texas) was implanted in the capsular bag. (The AcrySof is the IOL of choice in cases with a dubious IOL power calculation that may require subsequent removal). The Holladay formula for IOL calculation in postrefractive surgery eyes predicted an IOL power of 17.00 D to provide a neutral refractive result. The Camellin formula called for an IOL power of 18.83 D. My own formula predicted an IOL power of 24.50 D, which is what I chose to implant.

One hour after surgery, autorefraction showed a myopic shift with refraction of approximately -3.50 D; I decided to remove the lens and replace it with another, this time with a power of 20.50 D. This operation was problem-free; however, a minor complication occurred at the end of surgery. During hydration of the incision, two of the RK incisions opened internally. I sutured the incisions with 10-0 nylon and removed them 8 days after surgery. However, at that time, one of the incisions opened again, allowing a considerable quantity of aqueous humor to escape.

If this were your patient, how would you proceed?

HOW I PROCEEDED
Again, I resutured the RK incision. This suture was removed 3 months later. The day after surgery, the patient's UCVA was 20/40 and 20/20 with a refraction of 0.00 -1.50 X 100; his IOP was 16 mm Hg.

Three months later, the patient's UCVA was 20/16. He subsequently decided to undergo cataract surgery on his left eye; following a cataract operation performed in another center, his BCVA with a refraction of +3.50 -6.00 X 60 was 20/200.

Despite the fact that considerable progress has been made in IOL calculation formulas for eyes previously subjected to refractive surgery, there is still no formula that can give predictable results. None of the formulas used predicted the correct IOL power for this patient. This case showed that even 30 years after RK, the incisions may still have a tendency to open. Additionally, when placing sutures in the RK incisions, they must remain in place for a fairly long time.

Lucio Buratto, MD, is an ocular specialist at Centro Ambrosiano di Microchirurgia Oculare, Milan, Italy. Dr. Buratto states that he does not have any financial interest in the products or companies mentioned. He may be reached at e-mail: office@buratto.com; Web site: www.buratto.com.