To study the factors influencing the reliability (accuracy) of autorefractometry before and after laser in situ keratomileusis (LASIK) for myopia and myopic astigmatism.
Retrospective case series.
A total of 250 consecutive eyes (132 patients, mean age 37 years) were included from the Department of Ophthalmology, J.W. Goethe University, Frankfurt, Germany. The mean preoperative spherical equivalent of the subjective refraction (SR) was −6.59 diopters (D) (−1.38 to −15.13). The patient data were reviewed before and 1 month after LASIK, including SR, objective automated refraction (AR), and visual acuity. Subgroup analysis was performed with regard to the preoperative myopia and the excimer laser optical zone (OZ). The Holm-Sidak and Wilcoxon matched-pairs tests were used for statistical analysis.
The correlation coefficient between AR und SR is r = 0.98 before LASIK versus r = 0.79 afterwards ( P < .001). The mean difference between the spherical equivalents (DSE) is +0.13 ± 0.51 D preoperatively versus −0.30 ± 0.58 D after LASIK ( P < .001). With a small optical zone size (5.0–5.5 mm) the postoperative difference is −0.61 D, versus −0.36 D for OZ 5.6–6.0, versus −0.16 for OZ 6.1–7.0 mm. With a rising preoperative amount of myopia, the postoperative AR results become increasingly more myopic than the SR.
Following LASIK, autorefractometry is less accurate than in nonoperated eyes. The reliability of the AR is influenced by the OZ and the preoperative amount of myopia, with a small OZ and high myopia resulting in a greater difference between AR and SR and with the AR determining more myopic results.
Laser in situ keratomileusis (LASIK) is currently the most commonly performed procedure to treat refractive errors. The range of the available correction, the reliability and safety of the results, and the speed of the visual recovery after the surgery made the procedure a revolutionary breakthrough in ophthalmology in the 1990s.
Besides the visual acuity, the measurement of the subjective refraction is a key test in both the preoperative assessment and the postsurgical follow-up. Clinicians and technicians are supported by automatic refractometry devices in determining the subjective refraction. The reliability (accuracy) of automatic refractometry has been demonstrated by several comparative studies in healthy nonoperated eyes.
The refractive and structural changes of the cornea made by LASIK may lead to inaccurate measurements by devices that have been designed to measure parameters in “untouched” corneal conditions (eg, keratometry). Clinicians have observed inaccurate automated refraction in certain cases after lamellar corneal refractive procedures, while the majority of the results remained reliable.
In this study, we aimed to identify the factors that could potentially influence the reliability of autorefractometry after LASIK for the correction of myopia and myopic astigmatism. We theorized that the preoperative amount of myopia and/or the excimer laser optical zone size may affect the automatic refractometry results postoperatively.
Study Population and Instrumentation
This retrospective study included a consecutive series of 250 eyes of 132 patients who were subjected to conventional LASIK for correction of either their myopia or myopic astigmatism. The mean patient age was 37 years (median, 35; range, 18 to 61). The study population consisted of 48.5% men (64/132) and 51.5% women (68/132). The study was performed at the Department of Ophthalmology, Johann Wolfgang Goethe-University, Frankfurt, Germany. Inclusion criteria were suitability for LASIK correction of myopia or myopic astigmatism and the absence of other pathologic ocular conditions and relevant systemic diseases. Only eyes with a pre- and postoperative best spectacle-corrected visual acuity of ≥20/40 were included.
All participants were evaluated for refractive surgical correction after a thorough ophthalmologic examination, including subjective and objective refraction (automatic refractometry) and indirect ophthalmoscopy of the posterior segment and were found to be suitable candidates for the LASIK procedure. The fundus examination was done through dilated pupils and after measurement of subjective and objective refractions. All patients were reviewed 1 month postoperatively, including subjective and objective refraction and uncorrected and best spectacle-corrected visual acuity. The subjective refraction was measured under the same lighting condition as preoperatively. The target refraction was not emmetropia in every case and intended undercorrection of myopia was not an exclusion criterion.
All surgeries were performed by 2 experienced refractive surgeons (T.K., G.S.) using the Hansatome microkeratome (Bausch & Lomb Surgical, Munich, Germany) for creation of a corneal flap followed by excimer laser ablation with a Technolas C-Lasik 217Z laser (Technolas, Bausch & Lomb Surgical). The flap diameter was 8.5 or 9.5 mm. The excimer laser uses a Munnerlyn-type ablation pattern (Planoscan).
Assessment of Refraction
The subjective refraction was determined without cycloplegia by experienced residents who had been trained in the measurement of refraction thoroughly using a phoropter. For determination of the spherical component the maximum plus method was used. The cross-cylinder technique was applied for the measurement of cylinder. Even in cases of good uncorrected postoperative visual acuity (>20/25), a subjective refraction was performed and plus glasses were tried in cases of excellent visual acuity (25/20 or better).
Objective refraction was measured by an autorefractometry device (Topcon RM A-2300, Tokyo, Japan) without cycloplegia using a standardized protocol by experienced ophthalmic nurses. Five measurements of the eye were taken and the values were automatically averaged. Autorefractometry had been performed prior to measuring the subjective refraction.
Data Collection and Statistical Analysis
All data were collected and evaluated by the same people (A.M., W.W.) using Microsoft Excel (version 9.0), Sigma Stat (version 3.5; SPSS Inc, Chicago, Illinois, USA), and Sigma Plot (version 10.0; Systat Software Inc., San Jose, California, USA). Besides the descriptive statistics, the Holm-Sidak and Wilcoxon matched-pairs tests were used for statistical analysis. P values less than .05 were considered statistically significant.
The following parameters were taken from the patient records: age, gender, side, pre- and postoperative subjective and objective refractions, best-corrected visual acuity (BCVA), and excimer laser optical zone size.
The mathematical “difference between the spherical equivalents” of the autorefractor-derived and subjective refraction was calculated for all 250 pre- and postoperative data. A power vector analysis of the pre- and postoperative data was performed according to Thibos and Horner.
Furthermore, the same calculations were performed after the data had been subdivided twice into the following 4 groups: first, subdivision with regard to the preoperative myopia (spherical equivalent) greater than −9.01 diopters (D) (Group I), between −6.01 and −9.00 D (Group II), between −3.01 and −6.00 D (Group III), and up to −3 D (Group IV); and second, subdivision with regard to the excimer laser optical zone size 5.0 to 5.5 mm (Group A), 5.6 to 6.0 mm (Group B), 6.1 to 6.5 mm (Group C), and 6.6 to 7.0 mm (Group D).
The correlation coefficient (r) between the spherical equivalents obtained with autorefractometry and subjective refraction was 0.98 before LASIK versus 0.79 after surgery ( P < .001). The overall mean difference was +0.13 ± 0.51 D (median, 0.13; range, −2.00 to +1.75) preoperatively versus −0.30 ± 0.58 D after surgery (median, −0.25; range, −3.13 to +1.63) ( P < .001). The mean preoperative spherical equivalent determined by the subjective refraction was −6.59 D (median, −6.5; range, −1.38 to −15.13) and the mean cylindrical power was −1.10 (median, −0.75; range, 0 to −5.75). Automated refractometry delivered the following preoperative results: mean spherical equivalent −6.46 (median, −6.38; range, −1.50 to −15.13) with a mean cylindrical power of −1.20 (median, −1.00; range, 0 to −6.75). After the surgery, the mean spherical equivalent by subjective refraction was −0.32 (median, −0.25; range, −5.00 to +3.88) with a mean cylindrical power of −0.44 (median, −0.25; range, 0 to −2.25). Automated refractometry delivered the following postoperative results: mean spherical equivalent −0.62 (median, −0.63; range, −5.25 to +2.88) with a mean cylindrical power of −0.68 (median, −0.50; range, 0 to −2.75).
The differences between subjective refraction and automatic refractometry are summarized in Tables 1 and 2 and illustrated in Figures 1 and 2 with special consideration to the preoperative amounts of myopia and the used excimer laser optical zones. Table 3 summarizes the results of power vector analysis.
|r a||Difference b||P||r a||Difference b||P|
|All||250||−1.38 to −15.13||0.98||+0.13||<.001||0.79||−0.30||<.001|
|I||45||Greater than −9||0.91||+0.43||<.001||0.88||−0.58||<.001|
|II||94||−6 to −9||0.78||+0.13||.012||0.82||−0.25||<.001|
|III||82||−3 to −6||0.92||+0.06||.201||0.55||−0.24||.001|
|IV||29||up to −3||0.76||−0.06||.820||0.59||−0.22||.026|
|r a||Difference b||P||r a||Difference b||P|