Acta Ophthalmologica 2014

Endothelial cell loss and refractive predictability in femtosecond laser-assisted cataract surgery compared with conventional cataract surgery Therese Krarup, Lars Morten Holm, Morten la Cour and Hadi Kjaerbo Department of Ophthalmology, Copenhagen University Hospital Glostrup, Copenhagen, Denmark

ABSTRACT. Purpose: To investigate the amount of endothelial cell loss (ECL) and refractive predictability by femtosecond laser-assisted cataract surgery (FLACS) compared to conventional phacoemulsification cataract surgery (CPS). Methods: Forty-seven patients had one eye operated by FLACS and the contralateral eye operated by CPS (stop and chop technique). Both eyes had intraocular aspheric lenses implanted. Uncorrected distance visual acuity (UDVA), corrected distance visual acuity (CDVA), central corneal endothelial cell count and hexagonality with a non-contact specular microscope were assessed preoperatively, 1–3 days postoperatively and 3 months postoperatively. Results: Three days postoperatively, mean ECL was 249 cells/mm2 (SD  744) (9.1%) by FLACS and 235 cells/mm2 (SD  681) (8.2%) by CPS (p = 0.87). Three months postoperatively, mean ECL was 274 cells/mm2 (SD  358) (11.4%) by FLACS compared with 333 cells/mm2 (SD  422) (13.9%) by CPS, (p = 0.30) 3 months postoperatively, hexagonality was decreased by 1.8% (SD  30) by FLACS and by 1.4% (SD  13) by CPS, (p = 0.84). The mean absolute difference from the attempted refraction was 0.37 dioptres (D) (SD  0.33) by FLACS and 0.41 D (SD  0.42) by CPS (p = 0.56). Mean CDVA was 0.89 (0.3; 1.25) by FLACS and 0.93 (0.4; 1.25) by CPS at 3 months postoperatively (p = 0.36). Within both groups, 70% gained a CDVA of 6/6. Mean surgery time was 9.3 min (SD  1.9) by FLACS and 8.0 min (SD  1.9) by CPS, (p = 0.0018). Mean phaco energy was 3.78 U/S (SD  5.1) and 5.45 U/S (SD  4.6) (p < 0.0001) by FLACS and CPS, respectively. Conclusion: We found no significant difference in ECL and refractive predictability between FLACS and CPS 3 months postoperatively. Key words: cataract surgery – corneal endothelial cell loss – femtosecond-assisted cataract surgery – LensAR – phaco energy

Acta Ophthalmol. ª 2014 Acta Ophthalmologica Scandinavica Foundation. Published by John Wiley & Sons Ltd

doi: 10.1111/aos.12406

Introduction Cataract surgery is one of the safest and most common operations in ophthalmology. Complications are rare but can be serious and sight-threatening. Conventional phacoemulsification surgery

(CPS) has a high success rate. Therefore, new surgical approaches need to show significant improvements to replace CPS. The recent introduction of femtosecond laser-assisted cataract surgery (FLACS) has shown promising results (Moshirfar et al. 2011). This technique

combines a high-resolution anterior segment imaging system with a femtosecond laser (FL), which delivers short pulses (10 15 seconds) of energy at near-infrared wavelength, which can be focused at very specific depths within the anterior chamber. The laser can perform corneal incisions, anterior capsulotomy and fragmentation of the lens nucleus. The hope of the latter has been to reduce the amount of required phaco energy and thereby reduce the risk of complications and result in a better visual outcome. Early reports show promising benefits of FLACS over CPS, including increased precision and reproducibility of anterior capsulotomy, reduced phaco energy and decreased collateral tissue damage (Nagy et al. 2009; Palanker et al. 2010; Abell et al. 2012, 2013; Bali et al. 2012a,b; Takacs et al. 2012). The number of complications seen in FLACS has not surprisingly been shown to decrease as experience is achieved (Roberts et al. 2013). A symmetric capsulotomy is probably important as capsulorhexis has been found to have a direct relation to effective lens positioning (ELP) which is the effective distance from the anterior surface of the cornea to the lens plane (Cekic & Batman 1999). ELP plays a key role for accuracy in IOL power formulas (Norrby 2008). Thus, a difference of only 1 mm in IOL position leads to approximately 1.25dioptre change in refraction (Lakshminarayanan et al. 1986; Erickson 1990; Sanders et al. 2006). In this regard, FLACS with laser-assisted capsulotomy may result in more predictable refractive outcomes.

1

Acta Ophthalmologica 2014

Decreased phacoemulsification energy during CPS has shown less endothelial cell loss (ECL) postoperatively (Murano et al. 2008; Storr-Paulsen et al. 2008; Shin et al. 2009). As the phaco energy can be reduced when performing FLACS, it is believed that FLACS might result in less ECL compared with CPS. However, long-term impact of FLACS on the cornea is yet to be determined. The aim of this paper was to compare the ECL and the refractive outcome between conventional and femtosecond laser-assisted cataract surgery.

Materials and Methods This prospective clinical trial with a consecutive cohort of 47 patients was offered FLACS on one eye and CPS on the contralateral eye at the Department of Ophthalmology, Frederiksberg, University Hospital of Copenhagen, Denmark. The study was following the tenets of the Declaration of Helsinki and was reported to the Danish Data Protection Agency. All patients volunteered to be included in the trial, and an informed consent was obtained from all. Inclusion criteria were visually significant cataract of any type and degree and age older than 18 years. Exclusion criteria were history of corneal or intraocular surgery, severe dry eye, corneal scars, corneal dystrophy, history of herpes keratitis, signs of keratoconus, history of uveitis, pseudoexfoliation syndrome, uncontrolled glaucoma, visually significant maculopathy, lack of cooperation and tremor. Preoperatively

The patients underwent assessment to establish eligibility for undergoing FLACS and CPS. All patients were used as their own control and the following measurements were taken: corrected distance visual acuity (CDVA), autorefraction, IOL power calculation using the SRK/ T formula (IOLMaster; Carl Zeiss Meditec, Jena, Germany), applanation tonometry and slitlamp evaluation including corneal, lens, cataract grade and fundus status. Endothelial cell density and the percentage of hexagonal cells were analysed using a non-

2

contact specular microscope (SP 2000P; Topcon, Tokyo, Japan) with the IMAGE-NET imaging system (version 4.0; Topcon). Corneal endothelial photographs were taken preoperatively, 1–3 days postoperatively and 3 months postoperatively. During each visit, 1–3 photographs of each cornea were taken and analysed automatically by IMAGENET imaging system. The mean of the readings was calculated and marked as the final reading. Infiniti Vision System, Alcon uses Cumulative Dissipated Energy (CDE) as a value for phaco energy. This is calculated as (phaco time 9 average phaco power) + (torsional time 9 0.4 9 average torsional amplitude). (The factor 0.4 represents approximate reduction in heat dissipated at the incision as compared to conventional phaco). Surgical technique

All patients were operated by the same experienced cataract surgeon (HK). To evaluate whether FLACS was superior to CPS regarding ECL, the eye with most dense cataract was operated with FLACS and the eye with less cataract with CPS. Femtosecond surgery

laser-assisted

cataract

The laser procedure was carried out using LensAR laser system (Distributed and Marketed in Europe by Topcon). The laser procedure was initiated by docking of the laser using a noncontact, fluid-filled patient interface, enabling imaging of the anterior chamber. LensAR uses Augmented Reality – a proprietary method of acquiring biometric data, using ray tracing to create a 3D reconstruction. The images and treatment plan were confirmed before the laser treatment was started. The 5.0 mm capsulotomy was performed, followed by lens fragmentation. The laser was disconnected and the remainder of the surgery was performed as phacoemulsification (Infiniti Vision System; Alcon, Fort Worth, TX, USA), and lens removal was performed mainly as a ‘piece by piece’ technique. Cataract surgery

A 1 mm sideport was created with a keratome at 50-degree position, followed by instillation of 2 ml lidocaine

(10 mg/ml) and ophthalmic viscosurgical device (OVD) (DiscoVisc; Alcon). Then, the two-step ‘Dimple Down’, clear cornea main incision was fashioned with a 2.4 mm angled keratome. A 5.0 mm continuous curvilinear capsulorhexis was created using Utrata forceps. Phacoemulsification and irrigation/ aspiration (I/A) was performed (Infiniti Vision System; Alcon) using the ‘Stop and Chop’ technique. An Acrysof IQ Aspheric IOL (Alcon) was implanted. Statistics

All statistics were performed using Sigma plot for Windows. Endothelial cell data were analysed using the Wilcoxon signed rank test with matched samples. Statistical analysis was performed using a 2-tailed Student’s t test. A probability level < 0.05 was considered statistically significant, and all p values reported are two-sided. We analysed the relationship between the phaco energy used and the ECL by first performing a Pearson’s test, which indicated that we could demonstrate the correlation between ECL and phaco energy by a linear regression. We did so using a stepwise backwards analysis. We analysed the data using a linear regression model initially with two separate regression lines: one for CPS and one for FLACS. No statistically significant differences were found between the parameters for CPS and FLACS, so the model could be reduced to a common regression line for the two methods (F-test, p = 0.4). The model could be further reduced, as the intercept parameter was not statistically significantly different from zero (F-test, p = 0.14).

Results Originally 55 patients were included in the study. Eight patients were later excluded (14.5%), one patient received FLACS on both eyes, one patient was so satisfied after the first operation that she cancelled the second operation, another patient had back pain and could not co-operate with FLACS, one patient had a facial bone structure making FLACS docking impossible, one patient died in between the two surgeries, three patients suffered from

Acta Ophthalmologica 2014

complications and failed to follow postoperative controls. Complications were the following: one developed cystic macula oedema in both eyes, one had a posterior capsule rupture during CPS and underwent vitrectomy, and one patient developed pigment epithelial detachment after FLACS. The preoperative refractive status of the included eyes in the FLACS and CPS group was as follows: number of eyes with astigmatism higher than 2D was five versus four. Number of eyes with hyperopia more than 3D was 12 versus 10. Number of eyes with myopia more than 6 D was seven versus six. Thus, the refractive preoperative condition of the eyes in the two groups was comparable. Cataract grade was preoperatively 2.1 (SD  0.78) in the FLACS group and 1.6 (SD  0.53) in the CPS group (p = 0.0005). All of the 47 patients enrolled completed the 3 month follow-up. The mean phaco surgery time was 9.3 min (SD  1.9) by FLACS and 8.0 min (SD  1.9) by CPS, (p = 0.0018). In FLACS, the mean LensAR surgery time (including docking, suction and laser time) during FLACS was 3.04 min (SD  0.95); thus, total operation time for FLACS was average 12.07 min. Mean total CDE was 3.78 U/S (SD  5.1) by FLACS and 5.45 U/S (SD  4.6) by CPS, (p < 0.0001). Preoperative mean ECD was similar in the two groups: 2505 cells/mm2 (SD  365) by FLACS and 2503 cells/ mm2 (SD  385) by CPS (p = 0.96). At the 3-day control, mean ECL was 249 cells/mm2 (SD  744) (9.1%) by FLACS and 235 cells/mm2 (SD  681) (8.2%) by CPS (p = 0.87). At the 3month control, mean ECL was 274 cells/mm2 (SD  358) (11.4%) by FLACS compared with 333 cells/mm2 (SD  422) (13.9%) by CPS (p = 0.309). There was an expected initial loss of hexagonality at the 3-day control, and then, a rise in hexagonality at the 3-month control (Table 1). At the 3month control, the percentage of hexagonal cells was decreased by 1.8% (SD  30) by FLACS and 1.44% (SD  13) by CPS, (p = 0.84). Pearson’s test showed a statistically significant correlation between ECL and used phaco energy (p < 0.0001) and a R2 value of 0.4, indicating that

Table 1. Preoperative and postoperative values of ECD, endothelial hexagonality, SEQ, as well as ECL, refractive difference from intended refraction, surgery time and used phaco energy for FLACS and CPS. FLACS Cataract grade Mean preop ECD (cells/mm2) Mean 1–3 days postop ECD (cells/ mm2) Mean 1–3 days postop ECL Mean 3 months postop ECD Mean 3 months postop ECL Preop hexagonality (%) 1-day hexagonality (%) Three-month hexagonality (%) Mean hexagonality loss (%) Preop SEQ (D) Three month postop SEQ (D) Mean difference SEQ (D) Mean absolute difference from intended refraction (D) Surgery time (min) CDE (U/S)

CPS

2.1  0.78 2505  365 2409  499 249 2230 274 55 49 53 1.8 0.66 0.62 0.08 0.37

          

744 (9.1%) 536 358 (11.4%) 8.08 9.3 11 30 4.3 0.55 4.37 0.33

9.3  1.9 3.78  5.1

p

1.6  0.53 2503  385 2369  500 235 2170 333 55 49 53 1.44 0.12 0.71 0.54 0.41

          

681 (8.2%) 603 422 (13.9%) 13 12 10 13.3 3.86 0.62 3.89 0.42

8.0  1.9 5.45  4.6

p = 0.0005 p = 0.96 p = 0.54 p p p p p p p p p p p

= = = = = = = = = = =

0.87 0.28 0.30 0.75 0.96 0.90 0.84 0.084 0.49 0.45 0.56

p = 0.0018 p < 0.0001

FLACS, femtosecond laser-assisted cataract surgery; CPS, conventional phacoemulsification surgery; ECD, endothelial cell density; SEQ, spherical equivalent; D, dioptres; min, minutes; CDE, cumulative dissipated energy; ECL, endothelial cell loss.

other factors may have an impact on ECL. When comparing CPS and FLACS, we found no statistically significant difference in the relationship between phaco energy used and ECL for the two methods: CPS and FLACS (p = 0.39). When reducing the model to a common regression line for CPS and FLACS, we found no other causes for ECL (p = 0.14). However, we found a linear dependency of the ECL on the phaco energy used which was highly statistic significant (p < 0.0001). The coefficient describing the linear relationship between phaco energy and ELC was estimated to be 56.7D ([cells/mm2]/[U/s])/  4.6 (estimate  SE). A power calculation with difference in means at 55 cells/mm2, standard deviation at 350 cells/mm2 and a patient group of 47 results in a power of 0.171. Mean preoperative spherical equivalent was 0.66 D (SD  4.3) in the FLACS group and 0.12 D (SD  3.86) in the CPS group. When pairing the eyes, the mean preoperative difference in spherical equivalent was 0.54 D (SD  2.08). The absolute mean difference from the attempted refraction was 0.37 D (SD  0.33) by FLACS and 0.41 D (SD  0.42) by CPS, (p = 0.56). When subanalysing the hyperopic patients (12 versus 10) and the myopic patients

(seven versus six), we found the absolute mean difference from the attempted refraction was 0.49 D (SD  0.45) by FLACS and 0.55 D (SD  0.41) by CPS, (p = 0.93) in the hyperopic patients and 0.28 D (SD  0.21) by FLACS and 0.11 (SD  0.08) by CPS, (p = 0.17) in the myopic patients. At 1 – 3-day control, mean UDVA was 0.48 (0.1; 1) and mean CDVA was 0.72 (0.16; 1) by FLACS versus 0.53 (0.2; 1) and 0.75 (0.3; 1) by CPS. At 3month control, mean UDVA was 0.55 (0.1; 1) and mean CDVA was 0.89 (0.3; 1.25) by FLACS versus 0.56 (0.1; 1) and 0.93 (0.4; 1.25) by CPS. Both groups had 83% CDVA ≥ 6/7.5 and 70% had CDVA 6/6.

Discussion The aim of this study was to compare the amount of ECL and refractive predictability by FLACS compared to CPS. Decreased phacoemulsification energy during CPS has shown less ECL postoperatively (Murano et al. 2008; Storr-Paulsen et al. 2008; Shin et al. 2009). ECL after CPS has been described to be between 1.4 and 23% (Burkhard et al. 1996; Richard et al. 2008; Hugod et al. 2011; Storr-Paulsen et al. 2013), but few studies have examined the impact on ECL after FLACS. Takacs et al. (2012) compared

3

Acta Ophthalmologica 2014

4

Schaick et al. 2005). This might explain our high standard deviation and relatively high value of ECL. We found that the ECL could be described as a linear function of the phaco energy used, (Fig. 1). This is a confirmation of earlier studies (Burkhard et al. 1996; Storr-Paulsen et al. 2008; Shin et al. 2009; Takacs et al. 2012). In the FLACS group in our present study, we reduced the phaco energy CDE used by one-third compared with the CPS group (Fig. 2). This percentage is less than previous reports of 37–43% (Nagy et al. 2009; Palanker et al. 2010; Takacs et al. 2012; Abell et al. 2013). A reason for this is the fact that the eye with the most dense cataract was operated with FLACS, but learning curve effects are also believed to play a role. We found the ECL to be 18% higher in CPS eyes than in FLACS eyes. As this roughly corresponds to the 33% lower phaco energy used in the FLACS group, it is considered likely that a significant reduction in ECL could be obtained if the use of phaco energy could be reduced further. Based on the results in this study, we performed a power calculation for a future study. With our results (difference in the mean ECL between FLACS and CPS of 58 cells/mm2 and a stan-

dard deviation of this difference of 388 cells/mm2), a sample size of 354 individuals is needed to obtain statistical significance of this difference at the 0.05 level with a power of 0.8. However, we have shown that the ECL was strongly correlated to the phaco energy used (Fig. 1), and it seems possible to reduce the phaco energy when the surgeon is further in the learning curve. If the phaco energy can be reduced to around 40%, the ECL should be reduced to around 50%. In this case, 7 6

Phaco energy

ECL in two groups of 38 patients, one group undergoing FLACS and the other group undergoing CPS. They found an insignificant reduction of 123 cells/mm2 (4.4%) by FLACS and 299 cells/mm2 (10.5%) by CPS (p > 0.05). Another study compared 146 eyes: one eye underwent FLACS and the other CPS. They found a mean ECL of 7.9%  7.8% and 12.1%  7.3% 1 week after surgery by FLACS and CPS, respectively, and 8.1%  8.1% and 13.7%  8.4 3 months after surgery by FLACS and CPS, respectively (p < 0.001) (Conrad-Hengerer et al. 2013). Others have found a significant reduction in mean ECL when comparing 150 eyes undergoing FLACS with 143.8 cells/ mm2  208 (5.9%) to 51 eyes undergoing CPS with 224 cells/mm2  188 (9.2%) (p = 0.022) 3 weeks after surgery (Abell et al. 2013). Our study found a nonsignificant difference between ECL after FLACS (11.4%) compared with CPS (13.9%) (p = 0.29) 3 months after surgery, hence a 18% reduction in ECL by FLACS compared with CPS (Table 1). We used an automatic computerized method to analyse the endothelial cells and used the mean of three readings as a final result. This method is not used in other studies comparing cataract surgery and ECL. Most often 1–3 ophthalmologists blindly and manually analyse the photographs and use the mean value found (Burkhard et al. 1996; Storr-Paulsen et al. 2008, 2013; Takacs et al. 2012). One study used an automatic cell count of fifty images and used the mean value as the final reading (Conrad-Hengerer et al. 2013). The single-surgeon nature of our study strengthens our results, and our mean values in ECL are consistent with other findings, but our standard deviation is higher than previous findings (Burkhard et al. 1996; Storr-Paulsen et al. 2008, 2013; Takacs et al. 2012; Abell et al. 2013; Conrad-Hengerer et al. 2013). The gold standard for endothelial cell analysis is manually counting the cells with a counting box, but the recommended and most used technique is the semiautomated method with a corrected IMAGE-NET ECD (van Schaick et al. 2005). We used uncorrected IMAGE-NET ECD which have been showed to have a SD  349 cells and a higher measured ECD compared with other counting techniques (van

FLACS CPS

5 4 3 2 1 0

FLACS

CPS

Fig. 2. Mean used phaco energy by FLACS and CPS expressed as mean total cumulative dissipated energy CDE. CDE by FLACS was 3.78  5.1 and by CPS: 5.45  4.6 (p < 0.0001). FLACS, femtosecond laserassisted cataract surgery; CPS, conventional phacoemulsification surgery.

Fig. 1. The relationship between the phaco energy used and the ECL. We analysed the data using a linear regression model initially with two separate regression lines: one for CPS and one for FLACS. No statistically significant differences were found between the parameters for CPS and FLACS, so the model could be reduced to a common regression line for the two methods (F-test, p = 0.4). The model could be further reduced, as the intercept parameter was not statistically significantly different from zero (F-test, p = 0.14). The slope, that is, the linear dependency of the ECL on the phaco energy used was highly statistically significant (F-test, p < 0.0001). The coefficient describing the linear relationship between phaco energy and ELC was estimated to be 56.7 ([cells/mm2]/[U/s])/ 4.6 (estimate  SE). FLACS, femtosecond laser-assisted cataract surgery; CPS, conventional phacoemulsification surgery; ECL, endothelial cell loss.

Acta Ophthalmologica 2014

the expected difference between FLACS and CPS would be in the order of 150 cells/mm2. With a standard deviation of 388 cells/mm2, such a difference should be detectable at the 0.05 level and a power of 0.8 with a sample size of only 55 individuals. Mean surgery time was 9.3 min  1.9 versus 8.0 min  1.9 (p = 0.0018) in FLACS and CPS, respectively. The longer surgery time in FLACS is believed to be due to the learning curve described in other studies (Roberts et al. 2013). We experienced a more equal operation time after initial learning curve. One of the advantages of femtosecond lasers is the excellent precision, centration and reproducibility of the capsulotomy (Friedman et al. 2011; Kranitz et al. 2011). Literature indicates that when compared to a manual made capsulorhexis, laser-cut capsulotomies show significantly less deviation from intended diameter, better capsule–IOL overlap and improved IOL centration (Friedman et al. 2011; Kranitz et al. 2011; Abell et al. 2012; Roberts et al. 2013). This should lead to better refractive results (Lakshminarayanan et al. 1986; Erickson 1990; Sanders et al. 2006). Better precision of ELP is of uttermost importance when operating on hyperopic eyes to gain emmetropia, whereas myopic eyes are more tolerant. Thus, FLACS may have potential when operating hyperopic eyes. The preoperative refractive status of the included eyes in the FLACS and CPS group was comparable, thereby allowing meaningful comparison. We found an absolute mean difference from the attempted refraction of 0.37 D (SD  0.33) by FLACS and 0.41 D (SD  0.42) by CPS. In both the FLACS and CPS group, 19.1% was 0 D from intended refraction and 72.3% within 0.5 D from intended refraction (Figs 3 and 4). Roberts et al. (2012) compared the visual and refractive results of 113 eyes operated by FLACS with a control group who had undergone CPS. They found an absolute mean difference from the intended refraction of 0.29 D  0.25 D by FLACS and 0.31 D  0.24 D by CPS (p = 0.512). So far, the refractive results of FLACS compared with CPS have shown a trend towards better refractive predictability with a high-percentage ending at  0.25 D from attempted

refraction (Roberts et al. 2012; Abell et al. 2013); however, the results have not been statistically significant (Palanker et al. 2010; Roberts et al. 2012; Abell et al. 2013). In our study, we

found no difference between the two groups. Not all patients can co-operate with FLACS. There can be difficulties docking very deep set eyes. In our study,

Fig. 3. Attempted refraction versus achieved refraction. The figure shows that most eyes had an attempted SEQ of 0.5 D. The circles below the line indicate that the eyes achieved more myopia than attempted and vice versa. The mean absolute difference from the attempted refraction was 0.37 dioptres (D) (SD  0.33) by FLACS and 0.41 D (SD  0.42) by CPS (p = 0.56). SEQ, spherical equivalent; FLACS, femtosecond laser-assisted cataract surgery; CPS, conventional phacoemulsification surgery.

Fig. 4. Comparison of postoperative absolute SEQ 3 months after surgery between FLACS and CPS. 19% achieved attempted refraction by FLACS and 19% by CPS. Few patients ended up being hyperopic whereas 17% treated by FLACS ended with an SEQ of less than 1D compared with 25% treated by CPS. FLACS, femtosecond laser-assisted cataract surgery; CPS, conventional phacoemulsification surgery; SEQ, spherical equivalent.

5

Acta Ophthalmologica 2014

two patients were cancelled just prior to surgery; one due to back pain and the other due to abnormal facial bone structure. One other study has reported similar problems (Abell et al. 2012).

Conclusion We found a 33% decrease in used phaco energy CDE when performing FLACS compared with CPS, ECL to be significantly related to phaco energy use, a trend towards a decrease in ECL with FLACS and a trend towards a higher refractive predictability in FLACS compared with CPS. To evaluate the impact of FLACS on the endothelial cells, a larger randomized cohort study with blinded manual cell counting needs to be performed. In conclusion, we found that the performance of FLACS is not significant superior to CPS when an experienced surgeon is performing both procedures. One has to take into account though that in this study, FLACS was performed on the most dense cataracts. Along with the fact that the surgeries were performed early in the learning curve of FLACS, we expect that a second study will show significantly better results after FLACS.

References Abell RG, Kerr NM & Vote BJ (2012): Femtosecond laser-assisted cataract surgery compared with conventional cataract surgery. Clin Experiment Ophthalmol 41: 455–462. Abell RG, Kerr NM & Vote BJ (2013): Toward zero effective phacoemulsification time using femtosecond laser pretreatment. Ophthalmology 120: 942–948. Bali SJ, Hodge C, Chen S & Sutton G (2012a): Femtosecond laser assisted cataract surgery in phacovitrectomy. Graefes Arch Clin Exp Ophthalmol 250: 1549–1551. Bali SJ, Hodge C, Lawless M, Roberts TV & Sutton G (2012b): Early experience with the femtosecond laser for cataract surgery. Ophthalmology 119: 891–899. Burkhard D, Kohnen T, Jacobi F & Jacobi K (1996): Long-term endothelial cell loss following phacoemulsification through a

6

temporal clear corneal incision. J Cataract Refract Surg 22: 63–71. Cekic O & Batman C (1999): The relationship between capsulorhexis size and anterior chamber depth relation. Ophthalmic Surg Lasers 30: 185–190. Conrad-Hengerer I, Al JUBURI M, Schultz T, Hengerer FH & Dick HB (2013): Corneal endothelial cell loss and corneal thickness in conventional compared with femtosecond laser-assisted cataract surgery: three-month follow-up. J Cataract Refract Surg. doi: 10. 1016/j.jcrs.2013.05.033. Erickson P (1990): Effects of intraocular lens position errors on postoperative refractive error. J Cataract Refract Surg 16: 305–311. Friedman NJ, Palanker DV, Schuele G et al. (2011): Femtosecond laser capsulotomy. J Cataract Refract Surg 37: 1189–1198. Hugod M, Storr-Paulsen A, Norregaard JC, Nicolini J, Larsen AB & Thulesen J (2011): Corneal endothelial cell changes associated with cataract surgery in patients with type 2 diabetes mellitus. Cornea 30: 749–753. Kranitz K, Mihaltz K, Sandor GL, Takacs A, Knorz MC & Nagy ZZ (2011): Intraocular lens tilt and decentration measured by Scheimpflug camera following manual or femtosecond laser-created continuous circular capsulotomy. J Refract Surg 27: 558–563. Lakshminarayanan V, Enoch JM, Raasch T, Crawford B & Nygaard RW (1986): Refractive changes induced by intraocular lens tilt and longitudinal displacement. Arch Ophthalmol 104: 90–92. Moshirfar M, Hsu M & Churgin D (2011): Femtosecond laser-assisted cataract surgery: a current review. Middle East Afr J Ophthalmol 18: 285. Murano N, Ishizaki M, Sato S & Fukuda Y (2008): Corneal endothelial cell damage by free radicals associated with ultrasound oscillation. Arch Ophtalmol 126: 1–6. Nagy Z, Takacs A, Filkorn T & Sarayba M (2009): Initial clinical evaluation of an intraocular femtosecond laser in cataract surgery. J Refract Surg 25: 1053–1060. Norrby S (2008): Sources of error in intraocular lens power calculation. J Cataract Refract Surg 34: 368–376. Palanker DV, Blumenkranz MS, Andersen D et al. (2010): Femtosecond laser-assisted cataract surgery with integrated optical coherence tomography. Sci Transl Med 2: 58ra85. Richard J, Hoffart L, Chavane F, Ridings B & Conrath J (2008): Corneal endothelial cell loss after cataract extraction by using ultrasound phacoemulsification versus a fluidbased system. Cornea 27: 17–21.

Roberts TV, Lawless M, Chan CC, Jacobs M, Ng D, Bali SJ, Hodge C & Sutton G (2012): Femtosecond laser cataract surgery: technology and clinical practice. Clin Experiment Ophthalmol 41: 180–186. Roberts TV, Lawless M, Bali SJ, Hodge C & Sutton G (2013): Surgical outcomes and safety of femtosecond laser cataract surgery: a prospective study of 1500 consecutive cases. Ophthalmology 120: 227–233. Sanders DR, Higginbotham RW, Opatowsky IE & Confino J (2006): Hyperopic shift in refraction associated with implantation of the single-piece Collamer intraocular lens. J Cataract Refract Surg 32: 2110–2112. van Schaick W, Van Dooren BTH, Mulder PGH & Volker-Dieben HJM (2005): Validity of endothelial cell analysis methods and recommendations for calibration in Topcon SP-2000P specular microscopy. Cornea 24: 538–544. Shin YJ, Engler C, Kang J, Hasmi S, Jun A, Gehlbach P & Chuck R (2009): The Effect of Phacoemulsification Energy on the Redox State of Cultured Human Corneal Endothelial Cells. Arch Ophtalmol 27: 1–7. Storr-Paulsen A, Norregaard JC, Ahmed S, Storr-Paulsen T & Pedersen TH (2008): Endothelial cell damage after cataract surgery: divide-and-conquer versus phaco-chop technique. J Cataract Refract Surg 34: 996–1000. Storr-Paulsen A, Jørgensen JS, Norregaard JC & Thulesen J (2013): Corneal endothelial cell changes after cataract surgery in patients on systemic sympathetic a-1a antagonist medication (tamsulosin). Acta Ophthalmol. [Epub ahead of print]. Takacs AI, Kovacs I, Mihaltz K, Filkorn T, Knorz MC & Nagy ZZ (2012): Central corneal volume and endothelial cell count following femtosecond laser-assisted refractive cataract surgery compared to conventional phacoemulsification. J Refract Surg 28: 387–391.

Received on September 4th, 2013. Accepted on March 1st, 2014. Correspondence: Hadi Kjaerbo, MD Department of Ophthalmology Copenhagen University Hospital Glostrup Kongevejen 172 2830 Virum Denmark Tel: +45 20211929 Email: [email protected]