Journal of Refractive Surgery

Original Article Supplemental Data

Factors Influencing Cyclotorsion During Photorefractive Keratectomy

Fateme Alipour, MD; Fateme Veisi Hampa, MS; Elham Ashrafi, PhD; Shima Dehghani, MD

Abstract

PURPOSE:

To determine predictive factors for intra-operative cyclotorsion in photorefractive keratectomy (PRK).

METHODS:

A retrospective statistical analysis of medical records pertaining to 3,996 eyes undergoing PRK was conducted. Outcome measures of this study were the likely existence of statistically significant relations between diverse and potentially influential factors and the occurrence of intraoperative cyclotorsion.

RESULTS:

A total of 96% of examined medical records indicated some degree of cyclotorsion with the absolute mean intraoperative value of 1.38° ± 1.67° (range: 0° to 13.6°). Absolute mean cyclotorsion showed no significant correlation with age (r = 0.14, P = .37). Female patients had significantly higher degrees of cyclotorsion versus males (P < .001). Right and left eyes showed no significant difference in absolute mean torsion (P = .05). Higher diopters of refractive errors, hyperopia, more than 2.00 diopters of cylinder, high pulse numbers (r = 0.26), and use of the advanced personalized treatment ablation algorithm were all significantly related to higher degrees of torsion (P < .0001). Ablation depth (r = 0.13) and surgeon appeared to be dependent factors.

CONCLUSIONS:

Incidence of intraoperative cyclotorsion is high in eyes undergoing PRK and most of them will experience some degree of torsion. Several diverse factors (sex, refractive error diopters, hyperopia, high cylinder, pulse numbers, and ablation algorithm) were significant predictors for higher degrees of the observed torsion.

[J Refract Surg. 2018;34(2):106–112.]

Abstract

PURPOSE:

To determine predictive factors for intra-operative cyclotorsion in photorefractive keratectomy (PRK).

METHODS:

A retrospective statistical analysis of medical records pertaining to 3,996 eyes undergoing PRK was conducted. Outcome measures of this study were the likely existence of statistically significant relations between diverse and potentially influential factors and the occurrence of intraoperative cyclotorsion.

RESULTS:

A total of 96% of examined medical records indicated some degree of cyclotorsion with the absolute mean intraoperative value of 1.38° ± 1.67° (range: 0° to 13.6°). Absolute mean cyclotorsion showed no significant correlation with age (r = 0.14, P = .37). Female patients had significantly higher degrees of cyclotorsion versus males (P < .001). Right and left eyes showed no significant difference in absolute mean torsion (P = .05). Higher diopters of refractive errors, hyperopia, more than 2.00 diopters of cylinder, high pulse numbers (r = 0.26), and use of the advanced personalized treatment ablation algorithm were all significantly related to higher degrees of torsion (P < .0001). Ablation depth (r = 0.13) and surgeon appeared to be dependent factors.

CONCLUSIONS:

Incidence of intraoperative cyclotorsion is high in eyes undergoing PRK and most of them will experience some degree of torsion. Several diverse factors (sex, refractive error diopters, hyperopia, high cylinder, pulse numbers, and ablation algorithm) were significant predictors for higher degrees of the observed torsion.

[J Refract Surg. 2018;34(2):106–112.]

Cyclotorsion occurs when changing from a sitting to a supine position (static cyclotorsion)1,2 and also intraoperatively (dynamic cyclotorsion) in eyes undergoing LASIK and photorefractive keratectomy (PRK).3–5 average amount of absolute intraoperative cyclotorsion has been reported to be approximately 2° to 3°.6 It can exceed 10° in some patients,6,7 and instances of up to 24° have also been reported.4

Intraoperative cyclotorsion can negatively affect the refractive outcome and vision quality of patients undergoing refractive surgery.6 It leads to undercorrection and decreased precision for astigmatism correction.8 Up to 14% and 35% of undercorrection can theoretically be assumed to occur for 4° and 10° of torsional misalignment, respectively.9 The other problem is reduction of the efficacy of wavefront-guided surgeries for correction of higher order aberrations10 and induction of higher order aberrations after refractive surgery, possibly due to misalignment of the suggested wavefront map and suboptimal astigmatism correction.11 Cyclotorsion of greater than 2° has been shown to be able to produce significant postoperative aberrations.12

Compensation of torsional eye movements with iris registration technology results in lower amounts of residual cylindrical error and undercorrection.13–15 It has also been shown to be effective in decreasing the amount of unwanted aberrations.11,16

Limited and inconclusive data are available on factors influencing these movements, especially for PRK. Older age, female gender, and surgical factors such as higher pulse numbers have been reported to be associated with higher amounts of torsion,17,18 whereas contradicting results have been reported.19 Data on the evaluation of the individual effects of different types of refractive errors (ie, myopia, hyperopia, and astigmatism) or different severities of the mentioned errors on torsional movements are largely lacking. Data comparing the effect of different ablation algorithms on cyclotorsion are also limited.18

The aim of this study was to evaluate the effect of preoperative and intraoperative variables on intraoperative cyclotorsion during PRK. We also evaluated the specific effect of different subgroups of refractive errors and ablation algorithms on intraoperative cyclotorsion.

Patients and Methods

Patient Population

This retrospective study reviewed consecutive records of 3,996 patients (one eye per patient and alternating between left and right eyes) who had PRK between October 2011 and January 2015 using iris recognition technology at the refractive surgery department of Farabi Eye Hospital and Eye Research Center (Tehran University of Medical Sciences). The inclusion criterion was PRK with iris registration for correction of myopia, hyperopia, myopic astigmatism, and hyperopic astigmatism using either the Plano-scan, advanced personalized treatment (APT), or tissue-saving ablation profiles. Additionally, patients who had a history of previous refractive surgery and preexisting ocular disease were excluded from this study.

Retrospective Medical Data

The following data were extracted and used in this study: laterality of the eye, age, sex, refractive error type (dioptric value in terms of sphere and cylinder), surgical technique (Plano-scan, APT, and tissue-saving), intraoperative rotation (degree), ablation depth (microns), number of pulses, and surgeon (all according to the Technolas laser unit report for each eye).

Additionally, slit-lamp biomicroscopy, Goldmann applanation tonometry, and dilated fundus examination were available in the retrospective medical data reviewed and corresponding patients were excluded from this study in case of any sign of existing ocular disease.

Iris Registration and Aberrometry Method

The surgeries were performed by seven different surgeons using a Technolas 217z100 laser platform and Keracor software version 4.21 (Technolas Perfect Vision, St. Louis, MO) with advanced control eye tracking technology, which compensates for eye movements in six dimensions. The sampling rates and active tracking ranges are 240 Hz and 3 × 3 mm for x/y axes and 50 Hz and ±0.5 mm for z axis. The dynamic rotational system tracks and simultaneously adjusts the ablation pattern for the entire duration of the procedure and is independent of the x-y-z eye tracking. It uses a sampling rate of 25 Hz within a ±15° range (resolution dimension of dynamic torsional misalignment measurement = 0.1°).

APT is a wavefront-guided treatment. The wavefront map and iris image had been acquired for both eyes of all patients who had undergone APT at the same time with the same device (B & L Zywave; Bausch & Lomb, Rochester, NY). A minimum of three aberrometry readings per eye had been obtained to ensure reliability. Patients had been encouraged to blink and one drop of lubricant was used if necessary. Also, the head positioning and eye alignment were confirmed before measurements. All eyes were dilated with tropicamide 1% (two drops separated by a 5-minute interval) after measuring mesopic pupils. Scotopic aberrometry was performed using a similar protocol after 20 minutes, according to the manufacturer's instructions.

Measured Outcomes

Evaluated parameters were laterality of the eye, age, sex, treated refractive error type, intraoperative rotation degree, ablation depth, number of pulses, surgical technique (Plano-scan, APT, and tissue-saving), and the effect of the surgeon. Myopic patients were divided into three groups (mild < 3, moderate = 3 to 5.9, and severe > 5.9). Hyperopic patients were categorized in two different ways to enhance statistical analysis of data: three groups (mild < 3, moderate = 3 to 5.9, and severe > 5.9) and two subgroups of low (≤ 3) and moderate to high (> 3). Cylinder or astigmatism was divided into low (≤ 2) and high (> 2) subgroups, and spherical equivalent was divided into positive/hyperopic correction and negative/myopic correction subgroups. Intraoperative torsion was measured in this study and computed as absolute torsion, incyclotorsion (negative), and excyclotorsion (positive) in degrees. It was also characterized as mild (≤ 5°) and severe (> 5°). The cylinder axis was divided into three groups of with-the-rule (80° to 100°), against-the-rule (0° to 10° and 170° to 180°), and oblique (10° to 80° and 100° to 170°) astigmatism.

Statistical Analysis

Statistical analysis was performed with SPSS software (version 25; SSPS, Inc., Chicago, IL). Normality of data was investigated using the Kolmogorov–Smirnov test. The chi-square test was performed to compare categorical data. The continuous data were compared using one-way analysis of variance. Additionally, Tukey's honestly significant difference post hoc test was run to confirm where the differences occurred between groups if they had shown an overall statistically significant difference in group means. Factors with a significant effect (P < .20) on the amount of measured cyclotorsion were further analyzed by multivariate linear regression (absolute mean cyclotorsion) to determine their independent effect. All reported P values were two sided. The inferences were considered to be statistically significant when P values were less than .01. Cyclotorsion was computed as absolute cyclotorsion in all analyses. Patients with no cylindrical error were excluded from calculations related to cylinder. Eyes with no intraoperative cyclotorsion (154 eyes) were excluded from the calculations related to incyclotorsion or excyclotorsion and mild or severe torsion.

Results

Laterality and Demographic Data

The study included 3,996 eyes (1,997 right eyes and 1,999 left eyes). The male and female distribution was 1,524 (38.1%) and 2,472 (61.8%), respectively. The average age of patients was 28.9 years (range: 18 to 56 years).

Refractive Status (Spherical Equivalent and Type of Refractive Error)

Mean spherical equivalent was −3.65 ± 2.16 diopters (D) (range: −13.00 to +6.25 D). A total of 3,868 (96.8%) eyes had myopic ablation and 128 (3.2%) had hyperopic ablation. Tables 12 show the subgroup distribution in each group.

Refractive Spherical Error Subgroup Distribution

Table 1:

Refractive Spherical Error Subgroup Distribution

Refractive Cylindrical Error Subgroup Distribution

Table 2:

Refractive Cylindrical Error Subgroup Distribution

Intraoperative Cyclotorsion

The absolute mean intraoperative value of cyclotorsion was 1.38° ± 1.67° (range: 0° to 13.6°). A total of 3,804 (95.2%) patients had experienced 5° or less of torsion and 192 (4.8%) patients had experienced more than 5° of torsion. A total of 154 (3.9%) eyes showed no intraoperative cyclotorsion. Of 3,842 (96.14%) eyes with reported cyclotorsion (> 0°), 1,747 (43.7%) eyes had incyclotorsion and 2,095 (52.4%) eyes had excyclotorsion.

Relationship Between Intraoperative Cyclotorsion and Demographic Changes and Eye Laterality

Mean absolute cyclotorsion showed no significant correlation with the age of the patients (Pearson correlation = 0.14, P = .37). Additionally, the mean age was not significantly different between eyes having incyclotorsion or excyclotorsion (P = .10). Female patients had significantly higher degrees of cyclotorsion compared to male patients (1.54° and 1.26°, respectively, P < .001); this was significant in the regression analysis. There was no significant difference between the percentage of eyes having incyclotorsion or excyclotorsion between males and females (P = .28).

The right and left eye showed no clinically significant difference in absolute mean torsion (1.33 and 1.43, respectively; P = .049). However, more of the left eyes (5.5%, 110) had shown high cyclotorsion degrees (> 5°) compared to right eyes (4.1% ,82) (P = .02). The right eyes had experienced significantly more excyclotorsion (56.5%) compared to 48.4% in left eyes (P < .001).

Relationship Between Intraoperative Cyclotorsion and Refractive Errors

Hyperopic corrections showed significantly higher mean degrees of cyclotorsion (P < .0001). Eighteen percent (24) of eyes with hyperopic corrections had 5° or more of torsion compared to 4.3% (168) for myopic corrections (P < .0001) (Table 3).

Comparative Data for Absolute Mean Degree of Torsion Between Groups With Different Refractive Errorsa

Table 3:

Comparative Data for Absolute Mean Degree of Torsion Between Groups With Different Refractive Errors

Table A (available in the online version of this article) demonstrates differences between different refractive subgroups in having more than 5° of torsion. Additionally, eyes with high cylindrical errors demonstrated significantly larger amounts of torsion compared to eyes with low cylindrical errors (P < .0001) (Table 3).

Comparison of Eyes With > 5 Degrees of Torsion Between the Refractive Error and Ablation Algorithm Subgroups

Table A:

Comparison of Eyes With > 5 Degrees of Torsion Between the Refractive Error and Ablation Algorithm Subgroups

Eyes experiencing incyclotorsion versus excyclotorsion did not differ between different refractive subgroups.

The overall comparison between eyes with with-the-rule, against-the-rule, and oblique astigmatism showed no significant difference for the mean degree of absolute cyclotorsion (P = .97), mild/severe degrees of cyclotorsion, or the tendency for incyclotorsion or excyclotorsion.

Relationship Between Intraoperative Cyclotorsion and Surgical Factors

Ablation Depth and Total Pulses. Ablation depth and number of total pulses showed a weak positive correlation with mean absolute cyclotorsion (r = 0.13 and 0.26. respectively; P < .0001). Eyes with incyclotorsion had received slightly higher ablation depths and number of pulses (69.1 µm and 2,423) compared to eyes with excyclotorsion (67.15 µm and 2,331) or no cyclotorsion (62.45 µm and 2,055). The mean ablation depth and number of pulses were also significantly higher in patients with 5° or more of torsion (77.14 µm and 3,227.24 and 68.52 µm and 2,328.22 for eyes with low and severe torsion, respectively; P < .0001).

Ablation Profile. The mean absolute cyclotorsion was significantly higher in eyes that had APT (3.10 ± 2.39) compared to eyes that had tissue-saving (0.88 ± 0.85) and Plano-scan (0.87 ± 0.82) ablation (P < .001). However, no significant difference was seen between eyes undergoing tissue-saving and Plano-scan ablation (P = .96).

Additionally, significantly higher numbers of eyes with severe cyclotorsion were seen in the APT group (20.1%, 179 eyes) compared to the tissue-saving (0.5%, 11 eyes) and Plano-scan (0.3%, 2 eyes) groups (P < .0001) (Table A). The difference between eyes experiencing excyclotorsion versus incyclotorsion was 11.4% and 13.3% in the tissue-saving and Plano-scan groups, respectively, and 2.3% for the APT group (Table B, available in the online version of this article).

Comparative Data on Torsion Presence and Direction Between Different Ablation Algorithm Subgroups

Table B:

Comparative Data on Torsion Presence and Direction Between Different Ablation Algorithm Subgroups

Surgeon. Patients experienced significantly different amounts of torsion when surgery was performed by different surgeons, but surgeon effect became insignificant in the regression analysis.

Eyes With No Intraoperative Torsion

Gender, laterality, and surgeon showed no significant effect on intraoperative cyclotorsion. There was no significant difference between the spherical and cylindrical refractive subgroups or the different subgroups for cylindrical axis or ablation depth and total pulses between eyes with and without cyclotorsion. In eyes with compound astigmatism, the ablation algorithm was shown to be significantly different in eyes with and without cyclotorsion (Table C, available in the online version of this article); the difference was insignificant for the simple spherical error and simple cylindrical error groups.

Comparison of the Presence of Cyclotorsion Between Different Ablation Algorithms in Eyes With Compound Astigmatism

Table C:

Comparison of the Presence of Cyclotorsion Between Different Ablation Algorithms in Eyes With Compound Astigmatism

Factors Predicting the Absolute Amount of Intraoperative Cyclotorsion

Linear regression was performed to predict factors influencing the amount of cyclotorsion. Significant predicting variables were sex (P = .004), high cylinder (P < .0001), hyperopia (P < .0001), severity of the spherical error (P = .001), total pulse numbers (P < .0001), and APT ablation algorithm (P < .0001). There was no independent effect observed for ablation depth (P = .22) and surgeon (P = .03).

Discussion

In this study, we showed that almost 97% of patients having PRK experienced some degree of torsion during surgery. The absolute mean torsion was approximately 1° less compared to data obtained by Prakash et al.,15,18 Arba Mosquera and Verma,6 Chang,7 and Febbraro et al.1 Additionally, considering that the eye tracker used in this study could only record ±15° of torsion, it cannot be certainly said that the maximum amount of actual torsion has been 14.8°. Higher amounts of torsion may have happened intraoperatively that have resulted in stopping laser ablation. Interestingly, it seems that not all patients experience cyclotorsion during refractive surgery because 154 (3%) of the patients in this study had no intraoperative torsion. Zero degrees of torsion was also reported by Niu et al.19 in a population of 1,524 myopic and astigmatic patients. Although type or severity of refractive error did not seem to affect torsion during surgery, the current study showed that surgical algorithm can influence the latter. For example, cyclotorsion could be estimated to happen in almost all patients undergoing APT because only 1% of this group showed no torsion compared to approximately 5% in the Plano-scan and tissue-saving groups. Among eyes showing torsion, the majority of eyes (95.2%) had experienced small degrees of torsion (≤ 5). Although a minority of eyes (4.8%) tended to show larger amounts of torsion, these patients will be affected most by the negative effects of torsion on refractive outcomes.

Female gender was shown to be a strong independent factor for higher amounts of torsion in this study. Shajari et al.17 used LASIK (Keracor 217z excimer laser) on 179 eyes and reported that female gender can influence intra-operative cyclotorsion. Additionally, in a study on 275 patients undergoing LASIK (Technolas 217z100 platform), Prakash et al.18 showed that the absolute range of intraoperative cyclotorsion is higher in female patients (3.2 ± 1.5) compared to male patients (2.76 ± 1.46).

Age did not become an important factor influencing the amount of intraoperative cyclotorsion (r = 0.14, P =.37). This was not consistent with the results of Shajari et al.,17 who reported a positive relationship between torsion and age, and Prakesh et al.,18 who showed a partial but weak correlation (−0.16; P = .014). Both studies had a smaller sample size (179 and 275 eyes, respectively) and Shajari et al.17 only evaluated myopic astigmatic patients. Niu et al.'s19 study on 1,524 eyes that had undergone wavefront-guided LASIK (VISX STAR S4) failed to show any significant effect for either age or gender on intraoperative cyclotorsion. The latter study did not report the effect of different refractive errors (spherical and cylindrical) and only included myopic astigmatic eyes.

The higher mean absolute torsion in the left eye and the higher number of eyes with more than 5° of torsion (severe) in the left eye may have also been the result of patient fatigue because the left eye is usually the second eye to undergo surgery. This study also showed that right eyes had mostly experienced excyclotorsion, whereas left eyes mostly had incyclotorsion. Febbraro et al.1 also reported more excyclotorsion in the right eye (19% vs 12%) and more incyclotorsion in the left eye (36% vs 11%). The direction of cyclotorsion did not show significant differences in refractive or surgical subgroups. It can be speculated that direction is determined either randomly or mostly by anatomic factors (right and left eye) and is not related to refractive or surgical factors.

Blurred retinal image has been shown to enhance unwanted eye movements.3 Higher degrees of cyclotorsion in eyes with positive spherical equivalents were suggestive for the role of image blur in induction of torsion in hyperopic patients. Subgroup analysis data of the simple spherical error group confirmed higher degrees of cyclotorsion in hyperopic patients compared to myopic patients. Both myopic and hyperopic patients showed higher degrees of torsion as the severity of the refractive error increased, but the difference only became significant between mild and moderate to severe hyperopia. The pattern observed in the simple spherical error group was also seen in the compound spherocylinder group, regardless of the accompanying low or high cylindrical error, and the torsion increased by the increase of the spherical component was again more prominent in hyperopic patients. These findings could also be an attribute of the effect of image blur because myopic patients tend to have good near vision and, because the laser spot is placed a few centimeters from the eye, image blur will theoretically be less prominent for myopic eyes, even in high amounts of myopia.

Subgroup analysis in the eyes with simple cylindrical errors showed higher amounts of torsion for eyes with high cylindrical errors (small sample size may be the reason this did not become significant). Eyes with compound spherocylinder errors also showed approximately twice the torsion (in terms of degree; 2.32°) in eyes with high astigmatism compared to eyes with low astigmatism (1.16°). Astigmatism is also a known factor for blurred near vision. Prior studies have evaluated the effect of spherical equivalent on torsion degrees and reported no relationship between spherical equivalent and torsion degrees; however, their major limitation is that none have performed subgroup analyses for different levels of astigmatism or types of refractive errors.14,18,19 Moreover, two of them could not assess the effect of hyperopia because they had only included patients with myopic astigmatism.18,19 Longer duration of ablation in patients with higher refractive errors—either spherical or astigmatism—can also be suggested as a cause for loss of concentration and anxiety and subsequent higher intra-operative torsion in these subgroups.

Compared to Prakash et al.,18 the correlations observed in this study were stronger for the relationship between torsion and ablation depth (r = 0.13, P < .001) and of relatively similar strength for the number of total pulses (r = 0.26, P < .001). The latter study18 had also observed a weak correlation between absolute range of intraoperative torsion and depth of ablation (partial correlation = 0.02; P = .70) and amount of pulses (partial correlation = 0.39; P = 1.6 ×108). The stronger effect observed in this study may be due to the higher power of this study.

Differences seen in the amount of torsion between surgeons lost significance in the regression analysis, which was indicative that the effect was probably driven by the difference in ablation algorithms and refractive errors of the patients.

To our knowledge, this is the first study in the literature to evaluate the effect of different subgroups of refractive errors (in terms of being spherical or cylindrical and also in terms of severity) on the amount of torsion. Additionally, it is also the first study to compare torsion between three different ablation algorithms (tissue-saving, APT, and Plano-scan) and different surgeons. The mentioned characteristics along with the large sample size and heterogenic population made it possible for this study to evaluate the independent effect of each variable.

This study showed that the incidence of cyclotorsion is high in eyes undergoing PRK and most eyes will experience some degree of torsion. Female gender, higher diopters of spherical error, hyperopia, high cylindrical error, higher total pulse numbers, and APT ablation algorithm could be considered significant predictors for higher amounts of the observed torsion. Severe torsion (> 5°) also tends to happen more in eyes with high cylinder, higher total pulse numbers, and eyes undergoing APT. We suggest further studies in this regard. In addition to guiding the surgeon to identify high-risk patients who should have surgery using iris registration technology, it can also be used by manufacturers to incorporate more accurate tracking systems for high-risk patients, especially when the APT algorithm is being used.

References

  1. Febbraro JL, Koch DD, Khan HN, Saad A, Gatinel D. Detection of static cyclotorsion and compensation for dynamic cyclotorsion in laser in situ keratomileusis. J Cataract Refract Surg. 2010;36:1718–1723. doi:10.1016/j.jcrs.2010.05.019 [CrossRef]
  2. Prickett AL, Bui K, Hallak J, et al. Cyclotorsional and non-cyclotorsional components of eye rotation observed from sitting to supine position. Br J Ophthalmol. 2015;99:49–53. doi:10.1136/bjophthalmol-2014-304975 [CrossRef]
  3. Fea AM, Sciandra L, Annetta F, Musso M, Dal Vecchio M, Grignolo FM. Cyclotorsional eye movements during a simulated PRK procedure. Eye (Lond). 2006;20:764–768. doi:10.1038/sj.eye.6701994 [CrossRef]
  4. Neuhann IM, Lege BA, Bauer M, Hassel JM, Hilger A, Neuhann TF. Static and dynamic rotational eye tracking during LASIK treatment of myopic astigmatism with the Zyoptix laser platform and Advanced Control Eye Tracker. J Refract Surg. 2010;26:17–27. doi:10.3928/1081597X-20101215-03 [CrossRef]
  5. Narvaez J, Brucks M, Zimmerman G, Bekendam P, Bacon G, Schmid K. Intraoperative cyclorotation and pupil centroid shift during LASIK and PRK. J Refract Surg. 2012;28:353–357. doi:10.3928/1081597X-20120124-03 [CrossRef]
  6. 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]
  7. Chang J. Cyclotorsion during laser in situ keratomileusis. J Cataract Refract Surg. 2008;34:1720–1726. doi:10.1016/j.jcrs.2008.06.027 [CrossRef]
  8. Tomita M, Watabe M, Yukawa S, Nakamura N, Nakamura 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]
  9. Swami AU, Steinert RF, Osborne WE, White AA. Rotational malposition during laser in situ keratomileusis. Am J Ophthalmol. 2002;133:561–562. doi:10.1016/S0002-9394(01)01401-5 [CrossRef]
  10. Bueeler M, Mrochen M, Seiler T. Maximum permissible torsional misalignment in aberration-sensing and wavefront-guided corneal ablation. J Cataract Refract Surg. 2004;30:17–25. doi:10.1016/S0886-3350(03)00645-X [CrossRef]
  11. Wu F, Yang Y, Dougherty PJ. Contralateral comparison of wavefront-guided LASIK surgery with iris recognition versus without iris recognition using the MEL80 Excimer laser system. Clin Exp Optom. 2009;92:320–327. doi:10.1111/j.1444-0938.2009.00362.x [CrossRef]
  12. Guirao A, Williams DR, Cox IG. Effect of rotation and translation on the expected benefit of an ideal method to correct the eye's higher-order aberrations. J Opt Soc Am A. 2001;18:1003–1015. doi:10.1364/JOSAA.18.001003 [CrossRef]
  13. Tjon-Fo-Sang MJ, de Faber J-THN, 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]
  14. Ghosh S, Couper TA, Lamoureux E, Jhanji V, Taylor HR, Vajpayee RB. Evaluation of iris recognition system for wavefront-guided laser in situ keratomileusis for myopic astigmatism. J Cataract Refract Surg. 2008;34:215–221. doi:10.1016/j.jcrs.2007.09.022 [CrossRef]
  15. Prakash G, Agarwal A, Ashok Kumar D, Jacob S, Agarwal A. Comparison of laser in situ keratomileusis for myopic astigmatism without iris registration, with iris registration, and with iris registration-assisted dynamic rotational eye tracking. J Cataract Refract Surg. 2011;37:574–581. doi:10.1016/j.jcrs.2010.11.025 [CrossRef]
  16. Khalifa M, El-Kateb M, Shaheen MS. Iris registration in wavefront-guided LASIK to correct mixed astigmatism. J Cataract Refract Surg. 2009;35:433–437. doi:10.1016/j.jcrs.2008.11.039 [CrossRef]
  17. Shajari M, Buhren J, Kohnen T. Dynamic torsional misalignment of eyes during laser in-situ keratomileusis. Graefes Arch Clin Exp Ophthalmol. 2016;254:911–916. doi:10.1007/s00417-016-3309-y [CrossRef]
  18. Prakash G, Ashok Kumar D, Agarwal A, et al. 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]
  19. Niu XG, Xu M, He XY. Multivariate correlation analysis of eye cyclotorsion degree in corneal refractive surgery. Guoji Yanke Zazhi (Int Eye Sci). 2015;15:950–952.

Refractive Spherical Error Subgroup Distribution

SubgroupMild (< 3)Moderate (3 to 5.9)Severe (≥ 6)Total
Absolute spherical error1,736 (43.4%)1,784 (44.6%)379 (9.5%)3,899
Simple spherical error443
  Myopia (total 96.8%)146 (33%)234 (52%)49 (11%)
  Hyperopia (total 3.2%)4 (0.9%)10 (2.3%)0
Compound–low cylinder2,744
  Myopia (total 2,650 or 96.57%)1,101 (40.1%)1,283 (46.7%)266 (9.7%)
  Hyperopia (total 94 or 3.42%)50 (53.2%)38 (40.4%)6 (6.4%)
Compound–high cylinder712
  Myopia (total 570 or 80.5%)324 (45.5%)192 (27%)54 (7.6%)
  Hyperopia (total 142 or 19.42%)111 (78.2%)28 (19.7%)3 (2.1%)

Refractive Cylindrical Error Subgroup Distribution

SubgroupLow (≤ 2)High (> 2)Total
Simple cylindrical error29 (29.9%)68 (70.1%)97
Compound myopic astigmatism2,650 (82.29%)570 (17.7%)3,220
Compound hyperopic astigmatism94 (39.84%)142 (60.01%)236

Comparative Data for Absolute Mean Degree of Torsion Between Groups With Different Refractive Errorsa

Refractive SubgroupAbsolute Degree of Torsion (Mean ± SD)RangeP
Hyperopic/myopic

  Hyperopic correction2.9 ± 2.40 to 13.6< .0001
  Myopic correction1.3 ± 1.60 to 9.9

Simple spherical error

  Mild myopia0.9 ± 1.00 to 6.2> .010
  Moderate myopia0.7 ± 0.70 to 3.7
  High myopia1 ± 0.90.1 to 4.6

  Mild hyperopia0.8 ± 1.10.2 to 2.5< .0001
  Moderate/high hyperopia2.9 ± 2.80.1 to 7.7

Simple cylindrical error

  Low2 ± 3.00 to 13.6.30
  High2.6 ± 2.40 to 9.4

Compound spherocylinder

  Myopic astigmatism1.3 ± 1.60 to 13.2< .0001
  Hyperopic astigmatism2.9 ± 2.30 to 12.6

  Low cylinder–myopia1.1 ± 1.30 to 11.3< .0001
  Low cylinder–hyperopia2.5 ± 2.10 to 9.2

  High cylinder–myopia2.2 ± 2.20 to 13.2< .0001
  High cylinder–hyperopia3 ± 2.50 to 12.6

Comparison of Eyes With > 5 Degrees of Torsion Between the Refractive Error and Ablation Algorithm Subgroups

Spherical ErrorMild MyopiaModerate MyopiaHigh MyopiaMild HyperopiaModerate/High HyperopiaP

Simple100020< .0001
Compound5.83.41.914.920< .0001

Cylindrical ErrorLowHighP

Simple10.311.8.573
Compound313.5< .0001

Surgical AblationTissue SavingAdvanced Personalized TreatmentPlano ScanP

Total0.519.90.3< .0001

Comparative Data on Torsion Presence and Direction Between Different Ablation Algorithm Subgroups

Ablation AlgorithmNo TorsionExcyclotorsionIncyclotorsionP
Tissue-saving1131,245978< .0001
  % within ablation group5%53%42%
Advanced personalized treatment10435455
  % within ablation group1%48%51%
Plano-scan31415314
  % within ablation group4%55%41%
Total1542,0951,747

Comparison of the Presence of Cyclotorsion Between Different Ablation Algorithms in Eyes With Compound Astigmatism

Ablation AlgorithmNo TorsionWith TorsionP
Tissue-saving971,948< .0001
  % within ablation group5%95%
Advanced personalized treatment8810
  % within ablation group1%99%
Plano-scan24569
  % within ablation group4%96%
Total1293,327
Authors

From Eye Research Center, Farabi Eye Hospital, Tehran University of Medical Sciences, Tehran, Iran (FA, EA, SD); and the Department of Nutrition, School of Health, Qazvin University of Medical Sciences, Qazvin, Iran (FVH).

The authors have no financial or proprietary interest in the materials presented herein.

The authors thank Dr. M. Heidari for his assistance.

AUTHOR CONTRIBUTIONS

Study concept and design (FA); data collection (FVH); analysis and interpretation of data (FA, EA, SD); writing the manuscript (SD); critical revision of the manuscript (FA, FVH, EA); statistical expertise (FVH, EA, SD); administrative, technical, or material support (FA); supervision (FA)

Correspondence: Shima Dehghani, MD, Eye Research Center, Farabi Eye Hospital, Tehran University of Medical Sciences, Tehran, Iran. E-mail: shima. dehghani@outlook.com

Received: May 18, 2017
Accepted: November 09, 2017

10.3928/1081597X-20171128-02

Sign up to receive

Journal E-contents