Advanced cataracts distort OCT measurements

Surgeons should discard OCT retinal nerve fiber layer measurements in cataractous eyes when signal strength is less than 6, study author says.

Donald L. Budenz

Significant progressive cataracts may result in spurious retinal nerve fiber layer thickness measurements obtained with optical coherence tomography, according to a study in which advanced cataract reduced signal strength and caused artificially thin retinal nerve fiber layer measurements.

Clinicians should exercise extreme caution when interpreting peripapillary retinal nerve fiber layer (RNFL) measurements in eyes with significant cataracts, especially when the signal strength is less than 6, Donald L. Budenz, MD, MPH, the corresponding study author, said. Dr. Budenz wants physicians who are treating glaucoma to be aware that these OCT measurements are unreliable.

“They need to abandon that methodology for following patients,” he said.

Dr. Budenz advised clinicians against removing a cataract simply to increase OCT signal strength and ensure the accuracy of RNFL thickness measurements.

“Since OCT is sort of a secondary outcome measure of glaucoma progression and disease, we wouldn’t recommend removing a cataract just to improve the OCT signal strength and get a more firm measurement,” he said.

Study results were published in the Journal of Glaucoma.

Preop, postop measurements

The prospective study included 45 eyes of 45 patients; 22 eyes had cataracts only and 23 eyes had cataracts and glaucoma. Mean patient age in each group was 67.6 years and 75.2 years, respectively.

Patients underwent assessment of Snellen visual acuity, slit lamp examination of the anterior segment, IOP measurement with Goldmann applanation tonometry, fundus examination and visual field testing.

Patients underwent phacoemulsification and received an AR40 IOL (Allergan) or AcrySof SA60AT IOL (Alcon) implanted in the capsular bag.

Cross-sectional scans of peripapillary RNFL were obtained with the Stratus time-domain OCT (OCT 3, Carl Zeiss Meditec) using the Fast RNFL Thickness (3.4) protocol 0 to 155 days before cataract extraction and 27 to 127 days postoperatively.

Normal signal strength was indicated by a value of at least 6; low signal strength was indicated by a value less than 6. Dr. Budenz noted that the threshold of 6 conformed with the manufacturer’s recommended cutoff.

Signal strength influence

Results showed that postoperative RNFL thickness was 9.3% higher than preoperative measurements. The difference was statistically significant (P = .002).

Preoperative signal strength was 5.8 and postoperative signal strength was 7.2. The 24.1% increase was statistically significant (P < .001).

In 28 scans of 28 eyes with preoperative signal strength of 6 or higher, data showed no significant changes in postoperative signal strength or RNFL thickness.

Conversely, in 17 eyes with preoperative signal strength less than 6, results showed significant differences between preoperative and postoperative signal strength (P < .001) and RNFL thickness (P = .005).

Data showed no significant differences in the degree of change in RNFL thickness or signal strength between eyes with cataract only and those with cataract and glaucoma.

There was no significant relationship between changes in IOP and RNFL thickness. Preoperative IOP was 15.2 mm Hg and postoperative IOP was 14.5 mm Hg; the difference was statistically insignificant.

Lack of uniform standard

Determining an industry-wide standard for signal strength is not currently feasible, Dr. Budenz said.

“One of the confusing things is that there are probably seven to 10 companies making spectral-domain OCT instruments and they each use different algorithms for measuring the retinal nerve fiber layer,” he said. Each manufacturer could stipulate a different signal strength.

Even though spectral-domain OCT is gaining momentum as a tool for glaucoma-treating physicians, it is also susceptible to the influence of cataract in making RNFL thickness measurements, Dr. Budenz said.

“The OCT laser light is attenuated by cataract, no matter how many signals you are measuring,” he said. “So, the results should not change just because you are measuring with spectral-domain rather than time-domain OCT.”

Ultrasound biomicroscopy is not a viable alternative to OCT for making RNFL thickness measurements, Dr. Budenz said.

“The trouble with ultrasound is that you cannot get the 5 µm or so resolution to segment the layers,” he said. “OCT works with light rays being bounced off of interfaces in the retina. You just cannot get that with high-resolution ultrasound. We are always going to be dependent on some optical system.” – by Matt Hasson

Reference:

• Mwanza JC, Bhorade AM, Sekhon N, et al. Effect of cataract and its removal on signal strength and peripapillary retinal nerve fiber layer optical coherence tomography measurements. J Glaucoma. 2011;20(1):37-43.

• Donald L. Budenz, MD, MPH, can be reached at Bascom Palmer Eye Institute, 900 NW 17th St., Miami, FL 33136; 305-326-6384; fax: 305-326-6337.
• Disclosure: The Department of Ophthalmology, University of Miami Miller School of Medicine, receives unrestricted grant support for research and instruments on loan from Carl Zeiss Meditec.