Corneal biomechanical property evaluation appears useful in forme fruste keratoconus screening

A retrospective case-control study comparing data from preoperative evaluations with a non-contact applanation tonometer (Ocular Response Analyzer, Reichert) suggest that this device may be useful in helping to identify eyes with forme fruste keratoconus at risk for post-LASIK ectasia.

Key Points

San Francisco-A non-contact applanation tonometer (Ocular Response Analyzer, Reichert) appears to be a useful adjunct for screening of forme fruste keratoconus (FFKc) in patients seeking laser vision correction surgery, said Cédric Schweitzer, MD, at the annual meeting of the American Society of Cataract and Refractive Surgery.

The ability of the device to screen for FFKc was evaluated in a retrospective study including 55 patients who had developed unilateral keratoconus and were the cases, and 125 patients who demonstrated stable refraction and keratometry outcomes for 24 months postoperatively and served as controls. Patients were identified from the database of the Department of Ophthalmology, University Hospital Pellegrin, Bordeaux, France, where Joseph Colin, MD, is the chairman. All patients had undergone a comprehensive evaluation preoperatively with a clinical exam, non-contact applanation tonometry, and topography (Orbscan, Bausch & Lomb).

In comparing data for the pooled case and control groups, statistically significant differences existed between the keratoconus cases and controls in all evaluated tonometry parameters, and these differences remained in most comparisons made after stratifying the cases and controls into subgroups based on preoperative corneal pachymetry.

"Our study suggests the [tonometer] is useful as a preoperative screening tool to be used in addition to corneal topography and clinical evaluation," he continued. "However, long-term analyses are required to follow the biomechanical and topographic characteristics of corneas in this study."

The patients in the keratoconus group were considered to have unilateral disease with a normal fellow eye if the fellow eye had best spectacle-corrected visual acuity of 10/10 or better and a topography KISA index <60% without any suspicious topographic pattern. Subgroups created based on preoperative central corneal thickness (CCT) were defined as follows: group 1, <500 µm; group 2, 500 to 539 µm; group 3, 540 to 579 µm; and group 4, ≥580 µm. Eyes with a CCT <470 µm were excluded from the study.

"We stratified eyes by CCT since corneal hysteresis [CH] and corneal resistance factor [CRF] have been shown to correlate with corneal pachymetry," Dr. Schweitzer said.

CH, CRF evaluated

In addition to analyzing the two quantitative values provided by the tonometer, CH and CRF, the infrared and air pressure curves from the applanation also were evaluated in collaboration with the biomechanical engineering department at The Ohio State University, Columbus.

Both the mean CH and CRF values were significantly less in the cases versus controls, with the mean between-group differences being 1.2 mm Hg for CH and 1.9 mm Hg for CRF. For CH, mean CH values were significantly lower in CCT groups 1 and 2, and the difference approached statistical significance in group 3. In group 4, a relatively large difference was observed between cases and controls in mean CH values, but only two cases occurred in this CCT subgroup, and the difference compared with controls was not statistically significant. Consistent statistically significant differences existed in mean CRF values comparing each CCT subgroup.

"We also analyzed sensitivity thresholds of CH and CRF for discriminating between cases and controls, and those results suggested that CRF is the better of the two parameters," Dr. Schweitzer said.

The analyses of the air pressure curves showed that the FFKc corneas all were applanated faster and at lower pressures than the controls. Examination of the shape of the infrared signal curve showed that the case eyes were applanated less regularly and with a longer delay compared with the controls.