Journal of Refractive Surgery

Original Article Supplemental Data

Factors Associated With Ocular Cyclotorsion Detected by High-Speed Dual-Detection Eye Tracker During Single-Step Transepithelial Photorefractive Keratectomy

Soheil Adib-Moghaddam, MD; Saeed Soleyman-Jahi, MD, MPH; Salar Tofighi, MD; Ghazale Tefagh, MD; Samuel Arba-Mosquera, PhD; George Kontadakis, MD, PhD; George D. Kymionis, MD, PhD

Abstract

PURPOSE:

To determine demographic, surgical, and preoperative visual factors affecting the level of static and dynamic cyclotorsion (SC and DC) in eyes undergoing single-step transepithelial photorefractive keratectomy (TransPRK).

METHODS:

In this cross-sectional study, 386 patients with different types of refractive errors scheduled for single-step TransPRK were enrolled. A comprehensive profile of personal, optic, and visual characteristics of patients as surgical parameters were collected. Statistical modeling was used to explore correlates of SC and DC before and during the refractive procedure, respectively.

RESULTS:

There was no difference in various indices of SC and DC between right and left eyes. Lower age (beta = −0.20), uncorrected (beta = −0.99) and corrected (beta = −0.72) visual acuities, and keratometry (beta = −0.09) were significantly associated with higher ranges of DC during the laser ablation procedure. Worse mesopic contrast sensitivity (beta = 0.24) and higher (beta = 0.002, left eyes) or lower (beta = −0.002, right eyes) kappa locus also showed significant associations with a higher range of DC. In cases of SC before the procedure, amount (beta = 0.46) and axis (beta = −0.003) of the astigmatism demonstrated notable associations.

CONCLUSIONS:

Through comprehensive modeling, age, visual axis indices, visual acuity, and contrast sensitivity were found to be the main factors significantly associated with dynamic ocular globe cyclotorsions during single-step TransPRK. This could help identify eyes at higher risk of cyclotorsion and its potential complications in refractive surgery.

[J Refract Surg. 2018;34(11):736–744.]

Abstract

PURPOSE:

To determine demographic, surgical, and preoperative visual factors affecting the level of static and dynamic cyclotorsion (SC and DC) in eyes undergoing single-step transepithelial photorefractive keratectomy (TransPRK).

METHODS:

In this cross-sectional study, 386 patients with different types of refractive errors scheduled for single-step TransPRK were enrolled. A comprehensive profile of personal, optic, and visual characteristics of patients as surgical parameters were collected. Statistical modeling was used to explore correlates of SC and DC before and during the refractive procedure, respectively.

RESULTS:

There was no difference in various indices of SC and DC between right and left eyes. Lower age (beta = −0.20), uncorrected (beta = −0.99) and corrected (beta = −0.72) visual acuities, and keratometry (beta = −0.09) were significantly associated with higher ranges of DC during the laser ablation procedure. Worse mesopic contrast sensitivity (beta = 0.24) and higher (beta = 0.002, left eyes) or lower (beta = −0.002, right eyes) kappa locus also showed significant associations with a higher range of DC. In cases of SC before the procedure, amount (beta = 0.46) and axis (beta = −0.003) of the astigmatism demonstrated notable associations.

CONCLUSIONS:

Through comprehensive modeling, age, visual axis indices, visual acuity, and contrast sensitivity were found to be the main factors significantly associated with dynamic ocular globe cyclotorsions during single-step TransPRK. This could help identify eyes at higher risk of cyclotorsion and its potential complications in refractive surgery.

[J Refract Surg. 2018;34(11):736–744.]

Excimer laser keratorefractive procedures have been proved to be successful for the treatment of refractive errors. Recently, refined single-step transepithelial photorefractive keratectomy (TransPRK) has shown promising results in correcting different types1–5 of refractive errors and improving quantitative and qualitative visual functions in patients.1–4,6 Moreover, improving ablation profiles in new excimer lasers has increased precision and predictability of results. The SCHWIND AMARIS Total-Tech laser (SCHWIND eye-tech-solutions, Kleinostheim, Germany) is a new platform that has shown promising results in laser-assisted in situ keratomileusis (LASIK) for myopia with or without astigmatism such as correcting refractions using new solutions such as higher repetition rates of beams to reduce ablation time, detection of limbus to compensate for pupil shift under various light conditions simultaneously with pupil centroid shift compensation, and compensation for eyeball movements in various directions. These features help to target laser beams precisely on the cornea and reduce higher order aberrations (HOAs).

One of the major concerns regarding different modalities of refractive surgeries is residual astigmatism. Residual astigmatism can result from cyclotorsion.7 Cyclotorsion results from the eye rotational movement around the visual axis, and can be static or dynamic.8,9 Static cyclotorsion (SC) is induced by the vestibular system in the supine position, whereas the pre-operative measurements were done in upright position.9 Dynamic cyclotorsion (DC) takes place during refractive procedures around preoperative static torsion.10 Furthermore, cyclotorsion occurs around the visual axis, which differs from the ablation axis, thus producing a lateral displacement combined with cyclotorsion.9

Arba Mosquera et al.11 predicted an approximate residual astigmatic error of 35% in eyes with more than 10° of calculated cyclotorsion. Cyclotorsion could also induce HOAs.12 Compensation for both SC and DC improves refractive outcomes.13

Despite the significant technical advancements in automatic eye tracking systems, cyclotorsion and its effect on refractive outcomes remains an issue to be addressed. Determining preoperative factors that induce a wider range of DC can help decision making regarding refractive surgeries. Prakash et al.10 showed that DC depends on age, sex, and duration of ablation. However, they did not investigate the association of corneal and refractive factors with the range of DC during refractive procedures. Thus, we aimed to determine a more comprehensive profile of demographic, surgical, and preoperative visual factors that can help to predict cyclotorsion during single-step TransPRK.

Patients and Methods

Patients

In this cross-sectional study, patients with different types of refractive errors with or without astigmatism who underwent single-step TransPRK at Bina Eye Hospital were enrolled. Patients completed a written informed consent. The study protocol was in accordance with the tenets of the Declaration of Helsinki and institutional review board approval was obtained.

The exclusion criteria were concurrent ocular disease, previous corneal or ocular surgery, severe dry eye, and systemic diseases with ocular involvement. We also excluded patients with nervous system disorders that could interfere with their control of eye movements (eg, cerebellar or basal ganglia lesions). All of the eyes had successful SC correction (SCC) measurements. Patients were asked to discontinue wearing hard or soft contact lenses 4 weeks prior to surgery.

Patient Assessment

Uncorrected distance visual acuity (UDVA), corrected distance visual acuity (CDVA), refraction, keratometry and topography with Scout (Optikon 2000 SPA, Rome, Italy) and Orbscan (Bausch & Lomb, Rochester, NY), contrast sensitivity in photopic and mesopic conditions (M&S Smart System 20/20; M&S Technologies Inc., Niles, IL), corneal wavefront aberrometry (Keratron Scout Corneal Analyzer; Optikon 2000 SPA), and ocular wavefront aberrometry (ORK Wavefront Analyzer; SCHWIND eyetech-solutions GmbH) were assessed preoperatively.

All of the eyes underwent a refined protocol of single-step TransPRK by the same surgeon (SAdib-Moghaddam) as previously described.1,3,6 The SCHWIND AMARIS 500E laser (SCHWIND eye-tech-solutions GmbH) was used to perform the procedures. Patients were asked to fixate on a distant blinking light to reduce eye movements during the operation. Before and during the ablation, the eye tracking system compensated for eye movements with SCC and DC correction (DCC), respectively. All treatments were completed without a break.

Before starting the procedure, the eye tracker in the laser system took images of the iris in a supine position. It then searched for landmarks as an offset relative to the iris images taken during the pretreatment measurements in an upright position and calculated the SC angle. SCC measurement is not active throughout ablation. It is a one-shot snapshot of the position of the eye at one specific time before the ablation and helps in rotating the ablation profile to compensate for the cyclorotation that happens when a patient assumes a supine position. Thereafter, repetitive images taken during the treatment were compared with the image taken at the beginning of procedure (SC value), as a baseline, to calculate the DC angle. The cyclotorsion calculated was compensated with the rotation of the ablation profile (DCC).

The laser system produced values of SCC, DCC minimum, and DCC maximum for each eye. DCC minimum and maximum indicate maximum amounts of incyclotorsional and excyclotorsional compensation performed, respectively. Inward and outward torsions of the eye with reference to the SC value produced a negative amount as incyclotorsion, and a positive amount as excyclotorsion, respectively (Figure A, available in the online version of this article). Specular symmetry was considered in defining torsions in the left and right eyes. The system could not follow the 360° torsion and had some limitations in tracking cyclotorsion beyond 180°.

Directions of cyclotorsion in right and left eyes. The figure schematically illustrates directions of excyclotorsion and incyclotorsion in right (OD) and left (OS) eye. In right eyes, clockwise cyclotorsion indicates incyclotorsion and counterclockwise cyclotorsion indicates excyclotorsion. Conversely, clockwise and counterclockwise cyclotorsions in the left eye indicate excyclotorsion and incyclotorsion, respectively.

Figure A.

Directions of cyclotorsion in right and left eyes. The figure schematically illustrates directions of excyclotorsion and incyclotorsion in right (OD) and left (OS) eye. In right eyes, clockwise cyclotorsion indicates incyclotorsion and counterclockwise cyclotorsion indicates excyclotorsion. Conversely, clockwise and counterclockwise cyclotorsions in the left eye indicate excyclotorsion and incyclotorsion, respectively.

Definitions

DCC range was calculated by subtracting the signed DCC minimum value from the DCC maximum value. We also calculated absolute SCC, DCC minimum, and DCC maximum values. In most of the eyes, the visual axis (line connecting the fixation point to the fovea) was not the same as the pupillary axis (line passing through the center of pupil perpendicular to the cornea). The angle between these two axes is called the kappa angle.14,15 Kappa locus is the angular location of the visual axis with respect to the coronal plane of the cornea. It falls in one of the four 90° quadrants of the corneal plane and has a value between 0° and 360°. The four quadrants in each eye were defined as follows: in the right eye, quadrant 1 (0° to 90°) is upper nasal; quadrant 2 (90° to 180°) is upper temporal; quadrant 3 (180° to 270°) is lower temporal; and quadrant 4 (270° to 360°) is lower nasal. In the left eye, quadrant 1 is upper temporal; quadrant 2 is upper nasal; quadrant 3 is lower nasal; and quadrant 4 is lower temporal (Figure B, available in the online version of this article).

Angular map of kappa locus in right and left eyes. The figure schematically illustrates four quadrants in each eye where the angular location of kappa locus falls. In both eyes, kappa locus is dominantly located in the lateral semicircles of the corneal plane, which corresponds to quadrants 2 and 3 in the right eye and quadrants 1 and 4 in the left eye. Right eyes with their kappa locus located in the second quadrant have higher risk of dynamic cyclotorsional movements during the transepithelial photorefractive keratectomy surgery, compared to those with their locus located in the third quadrant. In left eyes, the location of kappa locus in the first quadrant is associated with lower risk of cyclotorsion, compared to the fourth quadrant. OD = right eye; OS = left eye; Q = quadrant; T = temporal; N = nasal; red shading = high risk of cyclotorsion; green shading = lower risk of cyclotorsion

Figure B.

Angular map of kappa locus in right and left eyes. The figure schematically illustrates four quadrants in each eye where the angular location of kappa locus falls. In both eyes, kappa locus is dominantly located in the lateral semicircles of the corneal plane, which corresponds to quadrants 2 and 3 in the right eye and quadrants 1 and 4 in the left eye. Right eyes with their kappa locus located in the second quadrant have higher risk of dynamic cyclotorsional movements during the transepithelial photorefractive keratectomy surgery, compared to those with their locus located in the third quadrant. In left eyes, the location of kappa locus in the first quadrant is associated with lower risk of cyclotorsion, compared to the fourth quadrant. OD = right eye; OS = left eye; Q = quadrant; T = temporal; N = nasal; red shading = high risk of cyclotorsion; green shading = lower risk of cyclotorsion

Ablation Procedure

Comprehensive demographic and optical characteristics of each patient were considered to individualize the target ablation profile.3 The eyes were irrigated with balanced salt solution. No alcoholic solution was administered for epithelial removal. A sponge with 0.02% mitomycin C was placed over the stromal bed for 25 seconds in eyes with myopia greater than 6.00 diopters (D) and for 5 to 25 seconds in hyperopic eyes.16,17 The eyes were then flushed with cold balanced salt solution. Ablation was done in a single continuous session. The ablation profile was centered on the pupil center when pupillary offset (the distance between pupil center and corneal vertex) was less than 0.35 mm. For greater pupillary offset values, corneal vertex was considered as the center of the ablation profile.

Statistical Analysis

We used the paired t test to compare different indices of cyclotorsion between right and left eyes. Thereafter, the Hosmer and Lemeshow approach18 for model building was used to fit multiple regression models of independent predictors for cyclotorsion. We included only one eye of each patient in model building. In brief, we first used screening simple regression models in original scale of variables to detect primary candidates; in this stage, a screening P value less than .20 was considered acceptable according to the approach. Then we fit the primary raw multiple models using the candidate variables spotted by simple regression models. To specify the final independent determinants, a backward elimination method was used to reach to the final multiple models. Plausible interaction terms between final determinants were also checked. If we found different results for a specific determinant factor between two eyes, separate results of both eyes were reported. We used Stata/SE software (version 11.1; Stata Corp LP, College Station, TX) for statistical analysis.

Results

A total of 386 patients (239 women, 147 men) were included, with a mean age of 30 ± 7.1 years. Mean spherical equivalent and astigmatism were −2.98 ± 2.49 and −1.02 ± 1.13 D, respectively (Table 1). There was no significant difference in components of cyclotorsion between right and left eyes (Table 2). Signed SCC, absolute SCC, and DCC range were comparable in the right and left eyes included in this study (P = .67, .46, and .43, respectively).

Demographic, Ophthalmic, and Visual Characteristics of Eyes Included in the Study (N = 386 Eyes)

Table 1:

Demographic, Ophthalmic, and Visual Characteristics of Eyes Included in the Study (N = 386 Eyes)

Comparison of Different Components of Cyclotorsion Between Right and Left Eyes Undergoing Single-step TransPRK

Table 2:

Comparison of Different Components of Cyclotorsion Between Right and Left Eyes Undergoing Single-step TransPRK

Multivariable linear analysis of absolute SCC showed that higher degrees of astigmatism (beta = 0.46, P ≤ .001) were significantly associated with greater absolute SCC before initiation of procedure. It also suggested that astigmatism axis (beta = −0.003, P = .045) and central corneal thickness (beta = −0.005, P = .03) were inversely correlated with absolute SCC values (Table 3).

Multivariable Linear Model for Determinants of Absolute Static Cyclotorsion Correction in Eyes Undergoing Single-step TransPRK

Table 3:

Multivariable Linear Model for Determinants of Absolute Static Cyclotorsion Correction in Eyes Undergoing Single-step TransPRK

Multivariable logistic analysis showed that patients with a higher astigmatism magnitude (odds ratio [OR] = 1.62, P = .002) were more likely to have an absolute SCC greater than four. Furthermore, eyes with astigmatism axis in the range of 135° to 180° (OR = 0.47, P = .04) had the lowest probability of having an absolute SCC greater than 4 (Table 4).

Multivariable Logistic Model for Factors Associated With Likelihood of Absolute Static Cyclotorsion Correction > 4 in Eyes Undergoing Single-step TransPRK

Table 4:

Multivariable Logistic Model for Factors Associated With Likelihood of Absolute Static Cyclotorsion Correction > 4 in Eyes Undergoing Single-step TransPRK

The results of multivariable linear analysis of DCC range indicated that higher preoperative UDVA and CDVA (beta = −0.99, P = .002 and beta = −0.72, P = .006, respectively) were markedly associated with lower degrees of DCC. Age (beta = −0.025, P = .04) and minimum simulated keratometry (beta = −0.09, P = .03) were also inversely correlated with DCC. Higher (beta = 0.0025, P = .03 in left eyes) or lower (beta = −0.0025, P = .04 in right eyes) degrees of kappa locus and worse mesopic contract sensitivity (beta = 0.24, P = .04) were shown to result in greater degrees of DCC (Table 5).

Multivariable Linear Model for Determinants of Dynamic Cyclotorsion Correction Range in Eyes Undergoing Single-step TransPRK

Table 5:

Multivariable Linear Model for Determinants of Dynamic Cyclotorsion Correction Range in Eyes Undergoing Single-step TransPRK

The kappa locus (Figure B) was dominantly located in quadrants 2 and 3 (99.5%) in the right eyes and in quadrants 1 and 4 (99%) in the left eyes. Therefore, we just included quadrants 2 and 3 in logistic model building of the right eyes and quadrants 1 and 4 in case of left eyes. A DCC value greater than 4 occurred at rates of 40% and 23.4% in right eyes with a kappa locus located in quadrants 2 and 3, respectively. However, these rates were 7.1% and 34% in the left eyes with a kappa locus located in quadrants 1 and 4, respectively.

The multivariable logistic model demonstrated that eyes with worse CDVA (OR = 2.78, P = .003) and UDVA (OR = 2.30, P = .01) and worse mesopic contrast sensitivity (OR = 1.68, P = .03) had a higher risk of having a DCC value greater than 4. Conversely, older patients (OR = 0.94, P < .001) and those with higher simulated keratometry values (OR = 0.88, P = .04) had a lower risk of having a DCC value greater than 4. Finally, compared to left eyes with a kappa locus in quadrant 1, those with a kappa locus in quadrant 4 were more likely to have a DCC value of greater than 4 (OR = 7.88, P = .04). In right eyes, location of kappa locus in the third quadrant was associated with lower risk of cyclotorsion (OR = 0.42, P = .05) (Table 6, Figure B).

Multivariable Logistic Model for Factors Associated With Likelihood of Dynamic Cyclotorsion Correction Range > 4 in Eyes Undergoing Single-step TransPRK

Table 6:

Multivariable Logistic Model for Factors Associated With Likelihood of Dynamic Cyclotorsion Correction Range > 4 in Eyes Undergoing Single-step TransPRK

Except kappa locus, the direction and strength of association for other determinant factors were not notably different between right and left eyes. Furthermore, subgroup analysis did not reveal a notable difference in associating factors of SCC and DCC between myopic and hyperopic eyes.

Discussion

In this study, we showed that worse preoperative visual acuity (UDVA and CDVA) is associated with higher values of DCC. Moreover, worse preoperative contrast sensitivity and younger age were shown to be related with higher values of DCC. We also found that right and left eyes with a kappa locus located in higher quadrants of the corneal plane were subject to lower and higher DCC, respectively. Unlike DCC, SCC showed associations only with preoperative astigmatism magnitude and astigmatism axis.

Cyclotorsion is considered to be a significant factor affecting predictability of astigmatism correction. Arba Mosquera et al.11 predicted an approximately 35% residual astigmatic error in eyes with more than 10° of calculated cyclotorsion. Furthermore, cyclotorsion could induce HOAs. Venter12 compared surgical outcomes in eyes undergoing LASIK for myopia with astigmatism with or without cyclotorsion error correction. Cyclotorsion control resulted in better astigmatism and HOA reduction. Tomita et al.13 showed improvement in astigmatism outcomes with combined SC and DC rather than DC alone in eyes having undergone LASIK for myopic astigmatism.

The deleterious effects of ocular globe cyclotorsion during laser ablation on postoperative residual astigmatism and HOAs are, therefore, marked. Consequently, identifying parameters capable of predicting the extent of these cyclotorsional movements could help delineate eyes at higher risk of suboptimal postoperative outcomes.

Prakash et al.10 investigated predictive factors for DCC during LASIK using the Technolas 217z100 platform (Technolas Perfect Vision, St. Louis, MO). Similar to our finding of lower values of DCC among older patients, they also suggested longer duration of ablation causes higher values of DCC due to less effective cooperation of the patients. They also showed associations between DCC and gender, which was not supported by our results. However, they did not find any correlations with spherical equivalent, SC, depth of ablation, and optical zone. They also did not investigate other possible predictive factors such as visual acuity and keratometry parameters and contrast sensitivity. In this study, we focused on visual parameters to determine factors predicting DCC to draw an accurate risk stratification plan before the surgery.

Arba Mosquera and Arbelaez8 reported that eye movements vary between cooperative and uncooperative patients. They detected smaller torsional movements in cooperative patients who had smaller eye movements and good fixation. Our findings emphasize the higher risk of cyclotorsion compensation, specifically in eyes with poor visual acuity. Worse visual acuity can lead to more difficulty in fixating on the distant blinking light, so earlier tiredness of eyes and less cooperation of the patients can be expected. This could lead to wider eye movements. This highlights the need for an alternative mechanism not dependent on the visual ability in such eyes with poor visual acuity to fix the eye during the procedure and reduce eye movements effectively. As a second parameter that determines visual function, astigmatism commonly affects the vision when looking at lines and shapes rather than points. Therefore, it probably does not impair the ability of the eye to focus on the blinking light during the surgery and cannot influence cooperation. As a result, there is no correlation between degree of astigmatism and DCC.

Lower preoperative contrast sensitivity led to higher odds of DCC greater than 4. As an object's contrast decreases, eyes with lower contrast sensitivity can detect it with lower spatial frequencies. The wider radius of a distant blinking light lowers the spatial frequency of the light. Therefore, we suggest that the distant blinking light should have a wider radius to be detectable by eyes with low contrast sensitivity. Better detection of a blinking light leads to better fixation of the eye and subsequently lowers DCC during the procedure.

Previous findings have introduced angle kappa as a factor that can cause decentration during refractive procedures14; however, its role in induction of DC has been unclear so far. We could not detect any significant association between angle kappa and DCC. Hashemi et al.19 reported a mean angle kappa value of 5.13° ± 1.50° in Iranian myopic eyes. Similarly, Qazi et al.20 reported a value of 5.0° ± 1.2°. Their findings are close to our findings (5.35° ± 2.96°).

We found a notable association between kappa locus and risk of DCC. Left eyes with a kappa locus located in the higher quadrant of the corneal plane underwent significantly higher DC compared to left eyes with their kappa locus located in the lower quadrant of the plane. This association was inverse in the right eyes (Figure B). To the best of our knowledge, there exists no literature significantly related to our findings. One possible explanation for these different associations in right and left eyes could be ocular dominance. We did not test ocular dominance in this study; however, previous studies on large sample sizes have reported a rate of ocular dominance in right eyes of up to 78%.21 When fixating on distant objects, the dominant eye is used and reaches more precise alignment with the sight axis.22,23 This could potentially contribute to this difference spotted in our study. Previous studies have demonstrated associations of ocular dominance with visual function and ocular structure.21,24,25 Yet, future studies focusing on neuro-optics and optical physics could unravel more clues underlying these findings. Regardless of the underlying mechanism, this finding could have clinical implications in more precise stratification of eyes based on cyclotorsion risk during surgery.

Arba Mosquera and Arbelaez8 calculated SCC angle with the same laser system we used in this study. The average SCC was 1° ± 4° and the maximum incyclotorsion and excyclotorsion were −6° and 7°, respectively. The absolute amount of SCC with the Technolas platform reported by Prakash et al.10 was 3.64° ± 2.79°. They compared absolute SCC values in studies of different racial populations and concluded that racial factors may determine the amount of SC. Our study was conducted in an Iranian population. The absolute amount of pre-ablation SCC noted among our patients was 2.61° ± 2.17°. The maximum incyclotorsion was −8.1° and the maximum excyclotorsion was 11°. The racial factor and the larger sample size of our study could be the reason for these differences between the SCC results. In this study, we found preoperative astigmatism magnitude and axis as predictors of SCC. SC occurs due to the change in the position of the vestibular system to fixate the image on the retina, but the mechanism of its induction by corneal and refractive factors is unclear. Excyclotorsion was the prominent direction in both left and right eyes in our study. Ciccio et al.26 and Chernyak27 also reported excyclotorsion dominance for both left and right eyes.

In patients with high risks of cyclotorsion, some specific considerations could be emphasized in the preoperative procedures supervised by the surgeon to minimize the risk. Most of the preoperative focus revolves around precise estimation and correction of the SC. The SC measured preoperatively actually serves as a reference. In case of accurate estimation of the SC, the DC movements of the eye during surgery are expected to be detected and corrected based on an appropriate reference. The following steps can enhance the accuracy of SC estimation: taking multiple shots and choosing the one with the best configuration (eg, low pupil offset, regular shape and expected size of the pupil); avoiding any tilt in head-to-body alignment of the patient at the time of SC acquisition, especially at SC values of greater than 10°; proper positioning of the patient's bed aligned to the camera; and instructing the patient to fixate at the center of the imaging camera.

In addition to the preoperative considerations, some intraoperative measures could further reduce the risk of DC. To improve the patient's cooperation, the surgeons could implement new platforms with higher repetition rates that shorten the duration of ablation and the time the patient has to fixate on the blinking light. In eyes with poor contrast sensitivity, a background could be added to the blinking light with a color that contrasts more with the blinking light.

Our results could be generalized to other surgical techniques. Most laser platforms that perform other laser-assisted ablative techniques use the same method to detect and compensate for the eye movements during procedures. In addition, an apparatus similar to the one used in this study could also be used for other techniques. Finally, all risk factors of cyclotorsional movements identified in this study are inherent to the patient. Thus, we expect that most of our findings could be extrapolated to other techniques.

In case of flap-related techniques, such as LASIK and laser epithelial keratomileusis (LASEK), there is one specific consideration for extra risks of torsional movements when attempting the maneuvers to separate and lift the flap. These induced torsions are transient and the eyeball tends to return to a more normal state soon after the maneuvers. Thus, the overall risk of cyclotorsion is not notably different from TransPRK. Nevertheless, this makes it imperative for surgeons performing LASIK and LASEK to consider this induced DC in addition to SC, to account for the movements that rotate the eye to a more neutral position.

The main limitation of this study was that we focused on finding predictive factors of cyclotorsional movements without considering lateral displacements. Although the existing literature is limited, lateral movements are clinically relevant to the suboptimal results seen postoperatively. Another possible limitation is that the laser system cannot track the torsional movements with absolute precision, and the amount of eye movements are not exactly equal to the compensation performed by the laser system. For the sake of simplicity, we considered the amount of cyclotorsional compensation by the laser system as a proxy of the actual cyclotorsional movements of the eyes.

We found that the absolute amount of DCC in eyes undergoing single-step TransPRK was notably associated with visual acuity and location of visual axis. Age, contrast sensitivity, and keratometry were also predictive factors. Eyes with worse visual acuity and specific angular locations of their kappa locus on the corneal plane are subject to higher DCC values. These findings suggest characteristics of eyes at higher risk of complications due to greater DC during laser ablation. More efficacious cyclotorsion compensation in these eyes can reduce induction of HOAs and residual astigmatism with resultant improvement in visual quality.

References

  1. Adib-Moghaddam S, Arba-Mosquera S, Walter-Fincke R, Soleyman-Jahi S, Adil-Aghdam F. Transepithelial photorefractive keratectomy for hyperopia: a 12-month bicentral study. J Refract Surg. 2016;32:172–180. doi:10.3928/1081597X-20160121-01 [CrossRef]
  2. Fadlallah A, Fahed D, Khalil K, et al. Transepithelial photorefractive keratectomy: clinical results. J Cataract Refract Surg. 2011;37:1852–1857. doi:10.1016/j.jcrs.2011.04.029 [CrossRef]
  3. Adib-Moghaddam S, Soleyman-Jahi S, Salmanian B, et al. Single-step transepithelial photorefractive keratectomy in myopia and astigmatism: 18-month follow-up. J Cataract Refract Surg. 2016;42:1570–1578. doi:10.1016/j.jcrs.2016.08.029 [CrossRef]
  4. Adib-Moghaddam S, Soleyman-Jahi S, Adili-Aghdam F, Arba-Mosquera S, Hoorshad N, Tofighi S. Single-step transepithelial photorefractive keratectomy in high myopia: qualitative and quantitative visual functions. Int J Ophthalmol. 2017;10:445–452.
  5. Aslanides IM, Padroni S, Arba Mosquera S, Ioannides S, Mukherjee A. Comparison of single-step reverse transepithelial all-surface laser ablation (ASLA) to alcohol-assisted photorefractive keratectomy. Clin Ophthalmol. 2012;6:973–980. doi:10.2147/OPTH.S32374 [CrossRef]
  6. Adib-Moghaddam S, Soleyman-Jahi S, Sanjari Moghaddam A, et al. Efficacy and safety of transepithelial photorefractive keratectomy. J Cataract Refract Surg. 2018;44:1267–1279. doi:10.1016/j.jcrs.2018.07.021 [CrossRef]
  7. Tjon-Fo-Sang MJ, De Faber J-T, Kingma C, Beekhuis WH. Cyclotorsion: a possible cause of residual astigmatism in refractive surgery. J Cataract Refract Surg. 2002;28:599–602. doi:10.1016/S0886-3350(01)01279-2 [CrossRef]
  8. Arba Mosquera S, Arbelaez MC. Use of a six-dimensional eye-tracker in corneal laser refractive surgery with the SCHWIND AMARIS TotalTech laser. J Refract Surg. 2011;27:582–590. doi:10.3928/1081597X-20110120-02 [CrossRef]
  9. Arba Mosquera S, Verma S. Effects of torsional movements in refractive procedures. J Cataract Refract Surg. 2015;41:1752–1766. doi:10.1016/j.jcrs.2015.07.017 [CrossRef]
  10. Prakash G, Ashok Kumar D, Agarwal A, Jacob S, Sarvanon Y, Agarwal A. Predictive factor analysis for successful performance of iris recognition-assisted dynamic rotational eye tracking during laser in situ keratomileusis. Am J Ophthalmol. 2010;149:229–237. doi:10.1016/j.ajo.2009.08.021 [CrossRef]
  11. Arba-Mosquera S, Merayo-Lloves J, de Ortueta D. Clinical effects of pure cyclotorsional errors during refractive surgery. Invest Ophthalmol Vis Sci. 2008;49:4828–4836. doi:10.1167/iovs.08-1766 [CrossRef]
  12. Venter J. Outcomes of myopic LASIK with and without NIDEK active torsion error correction. J Refract Surg. 2009;25:985–990. doi:10.3928/1081597X-20091016-03 [CrossRef]
  13. Tomita M, Watabe M, Yukawa S, Nakamura N, Makamura T, Magnago T. Supplementary effect of static cyclotorsion compensation with dynamic cyclotorsion compensation on the refractive and visual outcomes of laser in situ keratomileusis for myopic astigmatism. J Cataract Refract Surg. 2013;39:752–758. doi:10.1016/j.jcrs.2012.11.030 [CrossRef]
  14. Park CY, Oh SY, Chuck RS. Measurement of angle kappa and centration in refractive surgery. Curr Opin Ophthalmol. 2012;23:269–275. doi:10.1097/ICU.0b013e3283543c41 [CrossRef]
  15. Schwiegerling JT. Eye axes and their relevance to alignment of corneal refractive procedures. J Refract Surg. 2013;29:515–516. doi:10.3928/1081597X-20130719-01 [CrossRef]
  16. Shojaei A, Ramezanzadeh M, Soleyman-Jahi S, Almasi-Nasrabadi M, Rezazadeh P, Eslani M. Short-time mitomycin-C application during photorefractive keratectomy in patients with low myopia. J Cataract Refract Surg. 2013;39:197–203. doi:10.1016/j.jcrs.2012.09.016 [CrossRef]
  17. Adib-Moghaddam S, Soleyman-Jahi S, Tefagh G, Tofighi S, Grentzelos MA, Kymionis GD. Comparison of single-step transepithelial photorefractive keratectomy with or without mitomycin C in mild to moderate myopia. J Refract Surg. 2018;34:400–407. doi:10.3928/1081597X-20180402-02 [CrossRef]
  18. Hosmer D, Lemeshow S. Model-building strategies and methods for logistic regression. In: Hosmer D, Lemeshow S. Applied Logistic Regression, 2nd ed. Somerset, NJ: John Wiley & Sons; 2013:1–100.
  19. Hashemi H, KhabazKhoob M, Yazdani K, Mehravaran S, Jafarzadehour E, Fotouhi A. Distribution of angle kappa measurements with Orbscan II in a population-based survey. J Refract Surg. 2010;26:966–971. doi:10.3928/1081597X-20100114-06 [CrossRef]
  20. Qazi MA, Roberts CJ, Mahmoud AM, Pepose JS. Topographic and biomechanical differences between hyperopic and myopic laser in situ keratomileusis. J Cataract Refract Surg. 2005;31:48–60. doi:10.1016/j.jcrs.2004.10.043 [CrossRef]
  21. Ito M, Shimizu K, Kawamorita T, Ishikawa H, Sunaga K, Komatsu M. Association between ocular dominance and refractive asymmetry. J Refract Surg. 2013;29:716–720. doi:10.3928/1081597X-20130813-02 [CrossRef]
  22. Arba Mosquera S, Verma S. Bilateral symmetry in vision and influence of ocular surgical procedures on binocular vision: a topical review. J Optom. 2016;9:219–230. doi:10.1016/j.optom.2016.01.005 [CrossRef]
  23. Mapp AP, Ono H, Barbeito R. What does the dominant eye dominate? A brief and somewhat contentious review. Percept Psychophys. 2003;65:310–317. doi:10.3758/BF03194802 [CrossRef]
  24. Cheng CY, Yen MY, Lin HY, Hsia WW, Hsu WM. Association of ocular dominance and anisometropic myopia. Invest Ophthalmol Vis Sci. 2004;45:2856–2860. doi:10.1167/iovs.03-0878 [CrossRef]
  25. Jiang F, Chen Z, Bi H, et al. Association between ocular sensory dominance and refractive error asymmetry. PLoS One. 2015;10:e0136222. doi:10.1371/journal.pone.0136222 [CrossRef]
  26. Ciccio AE, Durrie DS, Stahl JE, Schwendeman F. Ocular cyclotorsion during customized laser ablation. J Refract Surg. 2005;21:S772–S774.
  27. Chernyak DA. Cyclotorsional eye motion occurring between wavefront measurement and refractive surgery. J Cataract Refract Surg. 2004;30:633–638. doi:10.1016/j.jcrs.2003.08.022 [CrossRef]

Demographic, Ophthalmic, and Visual Characteristics of Eyes Included in the Study (N = 386 Eyes)

ParameterValueaRange
Spherical equivalent (D)−2.98 ± 2.49−9.62 to 6.12
Astigmatism (D)−1.02 ± 1.13−5.25 to 0
UDVA (logMAR)0.55 ± 0.50−0.3 to 1.5
CDVA (logMAR)0.16 ± 0.41−0.3 to 1.5
DCC range (degree)2.93 ± 2.340 to 23.55
  < 4275 (71.24%)
  > 4111 (28.76%)
Age (year)29.86 ± 7.1016 to 55
CS photopic0.99 ± 0.710.40 to 8.0
CS mesopic1.11 ± 0.800.10 to 8.0
K1 (D)44.29 ± 1.5138.34 to 49.88
K1 axis (degree)96.34 ± 34.801 to 180
K2 (D)43.17 ± 1.4738.17 to 47.44
K2 axis (degree)68.51 ± 67.650 to 180
Central ablation (µm)129.24 ± 37.6131.02 to 227.16
Optical zone (mm)6.90 ± 0.424.49 to 7.8
Transitional zone (mm)1.38 ± 0.520.31 to 2.58
Temperature-start (°C)23.97 ± 1.4020.4 to 29.8
Temperature-end (°C)24.10 ± 1.3920.4 to 27.3
Humidity-start (%)27.94 ± 5.5714 to 48
Humidity-end (%)27.80 ± 6.2514 to 48
SimK maximum (D)44.79 ± 1.9040 to 49.3
SimK minimum (D)43.36 ± 2.0038 to 47.7
Pupil diameter (mm)4.36 ± 0.802.4 to 11.8
Kappa locus (degree)
  Right eye195.48 ± 17.79148.75 to 347.98
  Left eye313.19 ± 89.020.81 to 359.44
Kappa angle (degree)5.35 ± 2.972.04 to 9.75

Comparison of Different Components of Cyclotorsion Between Right and Left Eyes Undergoing Single-step TransPRK

ParameterTotalRight EyeLeft EyeP
Signed SCC0.39 ± 3.380.48 ± 3.350.34 ± 3.41.67
Absolute SCC2.61 ± 2.172.53 ± 2.442.69 ± 2.11.46
Signed DCC maximum1.91 ± 2.182.10 ± 2.331.74 ± 2.02> .99
Signed DCC minimum−0.92 ± 1.27−0.93 ± 0.15−0.90 ± 1.00.82
Absolute DCC maximum1.95 ± 2.152.10 ± 2.311.81 ± 1.96.16
Absolute DCC minimum1.05 ± 1.171.13 ± 1.3600.97 ± 0.95.13
DCC range2.90 ± 2.322.81 ± 2.182.98 ± 2.44.43

Multivariable Linear Model for Determinants of Absolute Static Cyclotorsion Correction in Eyes Undergoing Single-step TransPRK

ParameterAdjusted β (95% CI)P
Astigmatism (D)0.46 (0.30 to 0.61)< .001
Astigmatism axis (each 10º increments)−0.03 (−0.06 to −0.01).04
Central thickness (each 10-µm increments)−0.05 (−0.09 to −0.01).03

Multivariable Logistic Model for Factors Associated With Likelihood of Absolute Static Cyclotorsion Correction > 4 in Eyes Undergoing Single-step TransPRK

ParameterOR (95% CI)P
Astigmatism (D)1.62 (1.20 to 2.20).002
Astigmatism axis
  0 to 45Reference
  45 to 900.70 (0.35 to 1.36).29
  90 to 1350.64 (0.35 to 1.18).15
  135 to 1800.47 (0.23 to 0.96).04

Multivariable Linear Model for Determinants of Dynamic Cyclotorsion Correction Range in Eyes Undergoing Single-step TransPRK

ParameterAdjusted β (95% CI)P
UDVA−0.99 (−1.62 to −0.35).002
CDVA−0.72 (−1.24 to −0.21).006
SimK min (D)−0.09 (−0.17 to −0.005).03
Age (per 10-year increments)−0.2 (−0.5 to −0.11).04
Kappa locus (per 100º increments)
  When right eyes included−0.25 (−0.31 to −0.05).04
  When left eyes included0.25 (0.10 to 0.54).03
CS mesopic0.24 (0.11 to 0.59).04

Multivariable Logistic Model for Factors Associated With Likelihood of Dynamic Cyclotorsion Correction Range > 4 in Eyes Undergoing Single-step TransPRK

ParameterOR (95% CI)P
CDVA
  > 0.5Reference
  ≤ 0.52.78 (1.42 to 5.44).003
UDVA
  > 0.5Reference
  ≤ 0.52.30 (1.19 to 4.71).01
CS mesopic1.68 (1.06 to 2.66).03
SimK min0.88 (0.77 to 0.98).04
Age0.94 (0.91 to 0.97)< .001
Kappa locus (degree)a
  Lower quadrantReference
  Higher quadrant
    When right eyes included0.42 (0.16 to 0.98).05
    When left eyes included7.88 (2.66 to 50.96).04
Authors

From Bina Eye Hospital, Tehran, Iran (SAdib-Moghaddam, SS-J, ST, GT); TransPRK Research Center, Tehran, Iran (SAdib-Moghaddam, SS-J, ST, GT); Universal Council of Ophthalmology, Universal Scientific Education and Research Network, Tehran, Iran (SAdib-Moghaddam, SS-J, ST, GT, SArba-Mosquera); SCHWIND eye-tech-solutions, Kleinostheim, Germany (SArba-Mosquera); University of Valladolid, Valladolid, Spain (SArba-Mosquera); Institute of Vision and Optics, University of Crete, Greece (GK); and Jules Gonin Eye Hospital, University of Lausanne, Lausanne, Switzerland (GDK).

Dr. Arba-Mosquera is an employee of SCHWIND eye-tech-solutions and owns rights for a patent pertaining to SCHWIND eye-tech-solutions with no related payment. The remaining authors have no financial or proprietary interest in the materials presented herein.

AUTHOR CONTRIBUTIONS

Study concept and design (SAdib-Moghaddam, SS-J, ST, GT, SArba-Mosquera, GK, GDK); data collection (SAdib-Moghaddam, SS-J, ST, GT); analysis and interpretation of data (SAdib-Moghaddam, SS-J); writing the manuscript (SAdib-Moghaddam, SS-J, ST, GT, SArba-Mosquera, GK, GDK); critical revision of the manuscript (SAdib-Moghaddam, SS-J, ST, GT, SArba-Mosquera, GK, GDK); statistical expertise (SAdib-Moghaddam, SS-J, ST, GT, SArba-Mosquera, GK, GDK); administrative, technical, or material support (SAdib-Moghaddam, SArba-Mosquera); supervision (SAdib-Moghaddam, SArba-Mosquera, GK, GDK)

Correspondence: Soheil Adib-Moghaddam, MD, Bina Eye Hospital, Resalat Highway, Tehran 1634764651, Iran. E-mail: soheil.adibmoghaddam@gmail.com

Received: May 09, 2018
Accepted: September 24, 2018

10.3928/1081597X-20181001-01

Sign up to receive

Journal E-contents