Laboratory science Br J Ophthalmol: first published as 10.1136/bjophthalmol-2018-313421 on 4 April 2019. Downloaded from Comparison of anterior capsulotomy techniques: continuous curvilinear capsulorhexis, femtosecond laser-assisted capsulotomy and selective laser capsulotomy Sheraz Daya,‍ ‍ 1 Soon-Phaik Chee,2 Seng-Ei Ti,2 Richard Packard,3 David H Mordaunt4

►► Additional material is ABSTRACT also account for the majority of curable world published online only. To view Purpose To compare the anterior capsulotomy edge blindness2 and with improved small incision tech- please visit the journal online (http://dx.​ ​doi.org/​ ​10.1136/​ ​ tear strength created by manual continuous curvilinear niques the surgery is becoming safer and more bjophthalmol-2018-​ ​313421). capsulorhexis (CCC), femtosecond laser-assisted reliable. The burden of cataract surgery remains a capsulotomy (FLACS), and selective laser capsulotomy problem mainly in long-income and middle-income 1 Centre for Sight, East (SLC). countries however, with the ageing population in Grinstead, UK 2 Setting Singapore National Eye Centre, Singapore and industrialised nations, is likely to increase consid- Singapore National Eye Centre, Singapore, Singapore Excel-Lens, Livermore, California, USA. erably in the next two decades. Furthermore, with 3Arnott Eye Associates, London, Design Three armed study in paired human eyes. the advent of high performance multifocal and UK Methods Capsulotomies were performed in 60 cadaver 4 extended depth of focus lenses, lens extraction is Excel-Lens Inc, Livermore, eyes of 30 donors using CCC, Victus Femtosecond becoming increasingly popular as a means of vision California, USA Laser, (Bausch & Lomb, Rochester, New York, USA) or correction. Increasing volume, improvements in CAPSULaser, (Excel-Lens, Los Gatos, California, USA). Correspondence to efficiency, reproducibility and reliability of cataract Sheraz Daya, Centre for Sight, Three pairwise study groups each involved 10 pairs of surgery with reduced complications are motivating East Grinstead RH19 4RH, eyes. Study group 1: SLC eyes compared with fellow factors for further innovation. West Sussex, UK; ​sdaya@​ eyes with CCC. Study group 2: CCC eyes compared Modern cataract surgery is performed through centreforsight.com​ with fellow eyes with FLACS. Study group 3: FLACS eyes a small incision either by phacoemulsification or Presented in part at the compared with fellow eyes with SLC. whole lens removal through a capsular opening European Society of Cataract A shoe-tree method was used to apply load to the made by a continuous curvilinear capsulorhexis and Refractive Surgeons, Vienna, capsulotomy edge, and Instron tensile stress instrument (CCC).3 Over the last 10 years, the femtosecond September 2018. measured distension and threshold load applied to laser in combination with anterior segment imaging initiate capsule fracture. Relative fracture strengths and Received 23 October 2018 has been used to create a consistently circular Revised 6 January 2019 distension of CCC, FLACS and SLC were determined. capsulotomy of a defined size with good centra- Scanning electron microscopy (SEM) of capsule edges 4 5 Accepted 3 March 2019 tion. Additionally other devices have been devel- http://bjo.bmj.com/ were reviewed oped including a radio frequency cautery through Results Anterior capsulotomies behave as non-linear a disposable device containing a circular nitinol elastic (elastomeric) systems when exposed to an alloy ring placed in direct contact with the anterior external load. The pairwise study demonstrated that the capsule (Zepto Mynosys Cellular Devices, Fremont, SLC fracture strength was superior to that of CCC by a California, USA).6 7 A selective non-contact laser factor of 1.46-fold with SLC 277±38 mN versus CCC capsulotomy device has also been developed and

with 190±37 mN. Furthermore, CCC fracture strength on 5 April 2019 by Andrew Webb. Protected copyright. produces a capsulotomy of any size in 1 s and was superior to that of FLACS by a factor of 1.28-fold works by laser energy absorption of the trypan blue with CCC 186 + 37 mN versus FLACS 145 ± 35 mN (p dye stained anterior capsule (online supplementary < 0.001). This was determined by statistical analysis video). utilising the Wilcoxon matched-pairs signed-ranks test The ideal capsulotomy is one that can be and in accordance with the Consolidated Standards of performed rapidly and in a reproducible manner Reporting Trials guidelines. The capsule edge of SLC on with good centration on the crystalline lens, circular SEM demonstrated a rolled over edge anteriorly and an with good edge strength whereby there is little alteration of collagen. or no risk of radial anterior capsular tears during Conclusions The strength of the capsulotomy edge for SLC was significantly stronger than that of CCC which cataract surgery, lens prolapse and manipulation. © Author(s) (or their and both were significantly stronger than FLACS. The Radial anterior capsule tears can lead to intraocular employer(s)) 2019. No relative strengths can be explained by SEM of each type lens instability and at worst could during cataract commercial re-use. See rights surgery extend posteriorly to involve the posterior and permissions. Published of capsulotomy. by BMJ. capsule with subsequent vitreous loss in 50% of cases.8 9 To cite: Daya S, Chee S-P, Different methods of anterior capsulotomies Ti S-E, et al. Br J Ophthalmol Epub ahead of print: [please Introduction have different resulting metrics including the accu- include Day Month Year]. Cataract surgery is the most commonly performed racy of placement and centration as well as capsular doi:10.1136/ surgical procedure in the world with over 25 edge strength. Capsular strength is a factor that can bjophthalmol-2018-313421 million surgeries performed globally.1 Cataracts be influenced by a number of variables including

Daya S, et al. Br J Ophthalmol 2019;0:1–6. doi:10.1136/bjophthalmol-2018-313421 1 Laboratory science Br J Ophthalmol: first published as 10.1136/bjophthalmol-2018-313421 on 4 April 2019. Downloaded from method of capsulotomy used and capsulotomy size.10 We a guide for the size, circularity and centration of the CCC. The describe the comparative performance of three different tech- CCC was formed in a continuous manner using forceps under niques of capsulotomy specifically evaluating capsule strength OVD. Once completed the anterior capsule was rinsed with BSS and distension. and the capsulotomy disc removed with forceps.

Materials and methods Femtosecond laser assisted capsulotomy (FLACS) Three preclinical pairwise studies of ex vivo human cadaver eyes The FLACS technique was performed by rinsing the anterior were performed to compare relative threshold forces to initiate capsule with BSS followed by application of the OVD. A rigid an anterior tear for capsulotomies formed by selective laser contact lens and cohesive OVD were utilised to simulate the capsulotomy (SLC), CCC and femtosecond laser-assisted capsu- cornea and anterior chamber. The contact lens was centred on lotomy (FLACS) techniques. Specifically, the pairwise compari- the anterior capsule and the laser patient interface placed on it sons were divided into three study groups: Study group 1: SLC with hydroxypropyl methylcellulose (CAPSULgel, Excel-Lens, to CCC, Study group 2: CCC to FLACS and Study group 3: Los Gatos, California, USA). The eye was then docked to the FLACS to SLC. femtosecond laser (Victus, Bausch & Lomb, Rochester, New n all groups, the left and right cadaver eyes were randomly York, USA). The femtosecond laser was used at a pulse frequency allocated to the respective arms of the study. Pairs of phakic of 80 kHZ, pulse duration of 400–500 fs and near-infrared cadaver eyes from a donor in the specified age range of 35–70 wavelength of 1040 nm. The laser energy setting was 7.0 μJ years old were obtained within 72 hours postmortem from with spot and layer separations of 6 and 4 m, respectively. The either the SightLife Eye Bank (Seattle, Washington, USA) and µ capsulotomy size was set to an intended diameter of 5.0 mm. Singapore Eye Bank (SNEC, Singapore). Once the capsulotomy was accomplished, the patient interface In all eyes of all groups, to improve access and visibility as well was removed, and region of the capsulotomy was rinsed with as optical measurement of the capsulotomy, the cornea and iris BSS and the capsulotomy disc removed with forceps. was removed before performing the relevant procedure. In all eyes, a cohesive ocular viscoelastic device (OVD) (2% sodium hyaluronate acid, CAPSULVisc 2%, Excel-Lens, Los Gatos, Cali- Dimensional and stress evaluation analysis fornia, USA) was used. In all eyes, the lens nucleus was following The capsulotomy perimeter and circularity were measured for capsulotomy hydro-expressed using balanced salt solution (BSS) all study eyes using image processing software (provided by and resultant capsular bags filled with dispersive OVD (3% Excel-Lens, Los Gatos, California, USA). A microscope mounted sodium hyaluronate acid, CAPSULVisc 3%, Excel-Lens, Los camera captured images of the capsulotomy with an intraocular Gatos, California, USA). If cortex was observed it was removed ruler placed in the plane of the anterior capsule. The ruler was by automated irrigation and aspiration device (Stellaris, Bausch used to calibrate the dimensional scale. The capsulotomy rim & Lomb, Rochester, New York, USA). edge was defined by the reviewer identifying at least 40 points. The least-squares method was used to determined best-fit-circle Selective laser capsulotomy for these identification points: yielding the capsulotomy perim- The SLC technique involved initially rinsing the anterior capsule eter and centre. with BSS (Alcon, Fort Worth, Texas, USA). The cannula tip The methodology for determining the load-extension curves of the trypan blue syringe was placed centred on the anterior was with an Instron 3343 mechanical tester (Instron Corp, http://bjo.bmj.com/ capsule and 0.15 mL of trypan blue ophthalmic surgical solution Canton, Massachusetts, USA). The system was calibrated prior (CAPSULBlue, Excel-Lens, Los Gatos, California, USA) applied to use by the manufacturer. Measurements were performed using with a gentle pooling technique. This was left for 10 s and then a standardised method as described in detail by Daya et al10 In rinsed with 5 mL of BSS to remove any unabsorbed dye. The summary a uniquely designed wax-coated (to reduce probe fric- cohesive 2% OVD was then applied to the anterior capsule tion on the capsular edge) probe which exerted maximal force filling from one side in a continuous manner to the other, similar and extension on the capsular rim only was inserted into the on 5 April 2019 by Andrew Webb. Protected copyright. to the back-fill technique used with OVDs in surgery. The patient capsular bag through the capsulotomy with the rim located on interface lens was placed centred on the anterior capsule and the waist of the probe. The capsular bag supported by the probe OVD injected under the patient interface (CAPSULaser focusing was dissected from the globe with scissors cutting the zonules aid) to remove any air bubbles. The CAPSULaser focusing aid similar to the method described by Werner et al.11 The probes ensured that the laser was focused at the plane of the anterior were then connected to the Instron 3343 mechanical tester for capsule. A projected laser reticule allowed the location of the tensile strength testing. The system was reset to zero extension desired capsulotomy to be centred on the limbus or as appro- and the test commenced under computer control and video moni- priately selected by the surgeon. The capsulotomy diameter was toring. One arm of the fixture remained fixed in position while set to 5.0 mm in all cases. The laser footswitch was depressed the other was translated stretching the capsulotomy to its tear activating the laser to fire with one continuous pulse (rather than point. The capsular exterior of the capsular bag was frequently multiple pulses) for one second to create a 5.0 mm diameter irrigated with BSS during the procedures to prevent dehydra- capsulotomy. The patient interface was removed, and region of tion. Both stress and magnitude of extension were recorded for the capsulotomy was rinsed with BSS and the capsulotomy disc each increment of the stepper motor. The rupture point for stress removed with forceps. and extension were determined at the point of sudden drop in load. The threshold perimeter was calculated as twice the probe Continuous curvilinear capsulorhexis extension plus 4.5π for the curve probe perimeter. With the CCC technique, the anterior capsule was similarly Taking into consideration the influence of capsulotomy size rinsed with BSS and OVD applied as described above. The and continuity on threshold load and the incidence rate of ante- CAPSULaser device was also utilised to project a red laser rior tears10 the size and continuity was controlled in this study by circular reticule of 5.0 mm on to the anterior capsule to facilitate excluding pairs of eyes based on dimensional analysis. A pair of

2 Daya S, et al. Br J Ophthalmol 2019;0:1–6. doi:10.1136/bjophthalmol-2018-313421 Laboratory science Br J Ophthalmol: first published as 10.1136/bjophthalmol-2018-313421 on 4 April 2019. Downloaded from

Table 1 Donor age and gender distributions for each study group Study 1: SLC versus CCC Study 2: CCC versus FLACS Study 3: FLACS versus SLC Pair number Age (years) Gender Pair number Age (years) Gender Pair number Age (years) Gender 1 68 M 11 64 M 21 66 M 2 69 F 12 41 M 22 60 M 3 51 M 13 61 F 23 58 M 4 64 M 14 55 M 24 67 F 5 47 M 15 58 F 25 44 M 6 55 F 16 69 M 26 63 M 7 59 M 17 61 M 27 42 F 8 37 M 18 57 M 28 69 M 9 67 F 19 63 M 29 57 M 10 58 M 20 70 M 30 47 F Mean±SD 57±10 70% M Mean±SD 59±8 80% M Mean±SD 56±9 70% M CCC, curvilinear continuous capsulorhexis; FLACS, femtosecond laser-assisted capsulotomy; SLC, selective laser capsulotomy. eyes was rejected from analysis if one or both eyes demonstrated mN for SLC. The threshold perimeter for CCC at 25±2 mm one of the following three criteria: was also lower than SLC 29±2 mm as was extension at 5±1 1. Poor circularity as defined by non-circularity 4π⋅area of the mm for CCC versus 7±1 for SLC. In terms of perimeter stretch, best-fit-circle/perimeter distance from point to point2 being which is defined as the increase in perimeter relative to the less than 90%. original capsulotomy size, CCC had 59%±20% and SLC had 2. Poor circularity based on the method of minor and major 85%±15%. This indicates greater distensibility of the anterior axis ratio where the minor axis was less than 90% of the capsule following SLC compared with CCC (table 2). In Study major axis. group 2 where CCC was compared with FLACS, the tear load 3. Diameter sizing rejection criteria where the diameter was to cause a capsule tear was lower for FLACS (145±35 mN) than outside the range of the intended desired diameter of 5.0 CCC (186±37 mN) and the perimeter and perimeter stretch mm with a tolerance of ±0.2 mm. were slightly less for FLACS (23±2 mm, 46%±24%) than CCC The datasets were analysed to test superiority between the (24±2 mm, 53+22%) while extension was similar in both groups two arms for each pairwise study utilising the Wilcoxon Signed 12 at 5±1 mm. In Study group 3 comparing FLACS to SLC the Ranks Tests. The null hypothesis for each of the three pair- SLC treated capsules required a greater load (269±38 mN) than wise studies was that the mean difference including CI between FLACS (148±31 mN), along with larger extension perimeter pairs was zero, with the alternative hypothesis being the mean stretch (and perimeter) 78%±16% (28±1 mm) for SLC versus averages were different to the extent one arm is proven to be 46%±24% (23±1 mm) for FLACS. The load, perimeter and superior and the other arm is proven to be inferior (two-sided). extension for CCC in Study 1 and 2 were similar as was FLACS http://bjo.bmj.com/ in Study 2 and 3 and SLC in Study 1 and 3, demonstrating Scanning electron microscopy consistency of strength and distensibility between subgroups for The capsulotomy rim structures were evaluated for SLC, CCC each type of capsulotomy (table 2). and FLACS using scanning electron microscopy (SEM; Nano- To facilitate the Signed Rank Tests, the pairs were organised in craft, Los Gatos, California, USA). Anterior capsule tissue ascending absolute difference in perimeter stretch between the samples were collected for each capsulotomy technique. Some pairs for each of the three studies. SLC samples were also purposefully cross-sectioned by freezing In Study group 1 where SLC was compared with CCC 90% of on 5 April 2019 by Andrew Webb. Protected copyright. in liquid nitrogen and then sectioned with a stainless-steel the 10 pairs the perimeter stretch in SLC exceeded that of CCC, histology blade to investigate the nature of the capsulotomy edge with pair number 1 being the exception. For the load parameter, in more detail. The capsules were washed in Phosphate Buffered Saline for 60 min before being immersed in 1% aqueous solu- all 10 pairs of SLCs exceeded that of the CCCs. Using Wilcoxon tion, fixated in glutaraldehyde, then rinsed in distilled water, and Signed Ranks Test statistical analysis results indicated SLC was introduced into osmium tetroxide for 4 hours at room tempera- ture. Following this, the samples were dehydrated in increasing concentrations of ethanol. When the samples were critical point Table 2 Capsulotomy extensions, perimeters and loads for each dried they were sputter coated with 8–10 nm thick layer of gold pairwise study and examined using a scanning electron microscope (Vega 3, Tescan, Czech Republic). Extension Perimeter Perimeter Stretch Load (mm) (mm) (%) (mN) Results Study 1 SLC 7±1 29±2 85±15 277±38 The donor ages and genders for the three pairwise studies are CCC 5±1 25±2 59±20 190±37 reported in table 1. All three studies had similar age and gender Study 2 CCC 5±1 24±2 53±22 186±37 distributions with no meaningful differences. FLACS 5±1 23±2 46±24 145±35 The distributions for the capsulotomy extensions, perimeters Study 3 FLACS 5±1 23±2 46±24 148±31 and loads at the tear facture are reported in table 2, for the three SLC 7±1 28±2 78±16 269±38 pairwise studies. The mean load required to tear the anterior CCC, continous curvilinear capsulorhexis; FLACS, femtosecond laser-assisted capsule in Study 1 was 190±37 mN for CCC versus 277±38 capsulotomy; SLC, selective laser capsulotomy.

Daya S, et al. Br J Ophthalmol 2019;0:1–6. doi:10.1136/bjophthalmol-2018-313421 3 Laboratory science Br J Ophthalmol: first published as 10.1136/bjophthalmol-2018-313421 on 4 April 2019. Downloaded from

Figure 1 Pairwise comparison of SLC with respect to CCC for (A) perimeter stretch and (B) load. CCC, continuous curvilinear capsulorhexis; SLC,selective laser capsulotomy. statistically significantly higher for both parameters of perimeter micro-undulations resulting from repeated laser pulses with stretch and load (p≤0.01) (figure 1). some scattered pulses as evidenced by small holes adjacent to the In Study group 2 where CCC was compared with FLACS capsular edge (figure 4C). for perimeter stretch 70% of the 10 pairs demonstrate that the CCCs exceeded that of the FLACS with pairs 11, 12 and Discussion 13 being the exceptions (figure 2A). For the load, all 10 pairs Curvilinear capsulorhexis introduced in 1990 by Gimbel and for CCC exceeded those of the FLACS. The Wilcoxon Signed Nuehann3 was a revolutionary development in cataract surgery Ranks Test statistical analysis resulted in an inconclusive statis- and led to considerable improvement in safety and progress in tical significance (p value=0.2) when CCC was tested for superi- small incision phacoemulsification along with improved refrac- ority over FLACS with respect to perimeter extension. However, tive predictability with improved lens stability. However, ante- analysis of the load dataset demonstrated statistically significant rior capsule tear-outs can occur during performance of the superiority with the CCC requiring more load compared with procedure (primary anterior capsule tear) from a variety of FLACS to create an anterior tear, p value <0.01 (figure 2B). reasons including a shallow chamber, the early learning curve In Study group 3 where FLACS was compared with SLC, of a cataract surgeon13 or in complex situations like an intumes- all 10 pairs demonstrated statistically significant superiority cent cataract. This can lead to posterior extension and involve- for SLC for both increased perimeter stretch and load to tear ment of the posterior segment with vitreous prolapse. Anterior (p≤=<0.01) (figure 3). radial tears can also occur during the process of cataract surgery In summary, the pairwise studies demonstrated a statistically (secondary anterior capsule tear), from for instance prolapse of significant increase in perimeter stretch and load to tear for SLC a large hard nucleus during hydrodissection, instrument contact over both CCC and FLACS. Furthermore, CCC required statis- and even lens haptic contact and in turn also lead to posterior tically significant more load than FLACS to tear. extension. The rate of anterior capsule tear has been reported to http://bjo.bmj.com/ be between 0.79% and 5.5%.8 9 13 14 Posterior extension from an Scanning electron microscopy anterior capsule tear has been reported to be between 24%8 and Images of SEM are shown in figure 4. SLC (figure 4A) revealed 49%.9 A strong relatively tear resistant capsulotomy is therefore folding of the anterior capsule edge anteriorly. This appears to highly desirable. be from shrinkage of the anterior capsule and from local heat The ideal capsulotomy in addition to being relatively strong, generation from laser energy absorption of the trypan blue should be reproducible in size and also circular and central in stained capsule. order to adequately overlap the optic edge of the intraocular on 5 April 2019 by Andrew Webb. Protected copyright. SEM of CCC (figure 4B) showed a clean-cut edge revealing lens implant avoiding anterior optic capture with possible ante- the fibrillar collagen structure of the capsule. The femto- rior displacement and tilt which could lead to refractive error. second laser capsulotomy SEM showed a irregular edge with This has been the thrust of development of femtosecond lasers,

Figure 2 Pairwise comparison of CCC (black bars) with respect to FLACS (diagonal patterned bars) for (A) perimeter stretch and (B) load. CCC, continuous curvilinear capsulorhexis.

4 Daya S, et al. Br J Ophthalmol 2019;0:1–6. doi:10.1136/bjophthalmol-2018-313421 Laboratory science Br J Ophthalmol: first published as 10.1136/bjophthalmol-2018-313421 on 4 April 2019. Downloaded from

Figure 3 Pairwise comparison of FLACS (diagonal patterned bars) with respect to SLC (grey bars) for (A) perimeters stretch and (B) load. SLC,selective laser capsulotomy. radio frequency capsulotomies7 and SLC using a non-pulsatile dye Furthermore the study by Williams et al22 demonstrated also in absorbing laser. Circularity and centration of the femtosecond laser porcine eyes that larger capsules were significantly stronger, consis- has been well established and there have been reports of equivalent tent with a human eye study.10 strength to CCC by comparing the breaking force on human ante- This study compared three methods of anterior capsulo- rior capsule remnants removed at surgery.15 Previous studies on tomy, conventional CCC, femtosecond laser capsulotomy and a porcine eyes have shown greater strength of FLACS over CCC16–18 new development SLC using a non-pulsatile dye absorbing laser while another showed FLACS was weaker than CCC.19 The rough (CAPSULaser, Excel-Lens, Los Gatos, California, USA). Unlike capsular edge and discontinuity with tags of some FLACS capsu- previous studies which used porcine eyes, this study used human lotomies has been attributed to potential weakness and predispo- cadaveric eyes which were compared pairwise and equally matched sition to anterior tears. Several measures to improve capsule edge for age and cadaver time between each study group. The study also smoothness have been investigated including increasing vertical involved eliminating a pair of eyes where circularity and dimension spot separation.20 Additionally capsule openings created with criteria were not met as both these have been shown to influence higher energy levels were in porcine eyes found to be weaker extension and resistance to tear.10 18 The study revealed consis- than those created with intermediate to lower energy levels.21 22 tency of values between study groups for each type of capsulotomy (table 2) and this was achieved by reducing influencing variables and strictly controlling the use of eyes based on meeting dimension and circularity criteria. SLC was statistically more resistant to tear with a ratio of 146% compared with CCC. A result which was unexpected considering previous laboratory reports16–18 was the finding that CCC was statistically significantly more resistant to tear than FLACS where the ratio of FLACS to CCC was 78%. This http://bjo.bmj.com/ latter finding differs from the study by Thompson et al,7 which showed equivalence between CCC and FLACS, however the authors did not control for variables of circularity and dimensions with considerable variation in loads from eye to eye using the same capsulotomy method. Their study also had smaller numbers of eyes (six in number) in the group comparing CCC to femtosecond on 5 April 2019 by Andrew Webb. Protected copyright. laser capsulotomy which may have been a factor in not achieving statistical significance. Previous reports showing greater strength in FLACS were conducted in porcine eye studies16 18 which are known to have different anatomical thickness23 and elastic proper- ties from human eyes accounting for the different findings in this human eye study. In this study trypan blue stain was used in the SLC sub-group where a dye was required for uptake of energy from the non-pul- satile laser. There has been controversy over the impact of trypan blue stain on capsule fragility with Dick et al demon- strating increased fragility24 and Jaber et al later demonstrating no influence.25 If trypan blue does indeed increase fragility, the capsule edge of the SLC group was still statistically significant more resistant to tear than both CCC and FLACS. SEM of the CCC and FLACS was similar to that previ- ously reported26–29 with a very smooth anterior capsular edge with no irregularity on CCC, whereas FLACS showed consid- Figure 4 Scanning electron microscopy of selective laser capsulotomy erable irregularity of the anterior capsular edge with valleys (4A), continuous curvilinear capsulorhexis (4B) and femtosecond laser- and grooves with micro-undulations and a postage stamp like assisted capsulotomy (4C). appearance reflective of the repetitive pulsatile laser disruption

Daya S, et al. Br J Ophthalmol 2019;0:1–6. doi:10.1136/bjophthalmol-2018-313421 5 Laboratory science Br J Ophthalmol: first published as 10.1136/bjophthalmol-2018-313421 on 4 April 2019. Downloaded from

Patient consent for publication Not required. Table 3 Overall comparison of SLC, CCC and femtosecond laser- assisted capsulotomy with percentages relative to CCC. Provenance and peer review Not commissioned; externally peer reviewed. SLC CCC FLACS References Perimeter Relative to initial 78%±16% 53%±22% 46%±24% 1 Rao GN, Khanna R, Payal A. The global burden of cataract. Curr Opin Ophthalmol stretch capsulotomy 2011;22:4–9. Relative to CCC 120%±10% 100% 95%±10% 2 Resnikoff S. The global burden of eye disease: can ophthalmology meet the public need? 2013. Available: http://www.​icoph.org/​ ​downloads/GlobalBurdenandHRgap-​ ​ Load (mN) 273±37 186±40 146±32 Resnikoff.pdf​ Relative to CCC 146%±20% 100% 78%±20% 3 Gimbel HV, Neuhann T, Development NT. Development, advantages, and methods CCC, continous curvilinear capsulorhexis; FLACS, femtosecond laser-assisted of the continuous circular capsulorhexis technique. J Cataract Refract Surg 1990;16:31–7. capsulotomy; SLC, selective laser capsulotomy. 4 Mamalis N. Femtosecond laser: the future of cataract surgery? J Cataract Refract Surg 2011;37:1177–8. of the capsule edge. SLC SEM on the other hand, demonstrated 5 Nagy Z, Takacs A, Filkorn T, et al. Initial clinical evaluation of an intraocular anterior folding of anterior capsule resulting in the capsule edge femtosecond laser in cataract surgery. J Refract Surg 2009;25:1053–60. 6 FDA. 510K Premarket approval K170655 Mynosys cellular devices, Inc, 2017. at the convexity of the fold (figure 4A). Additionally there was Available: https://http://www.​accessdata.fda.​ ​gov/cdrh_​ ​docs/pdf17/​ ​K170655.pdf​ contracture of collagen resulting from laser energy absorption 7 Thompson VM, Berdahl JP, Solano JM, et al. Comparison of manual, femtosecond and localised thermal change. The finding of a statistically signif- laser, and precision pulse capsulotomy edge tear strength in paired human cadaver icant increased resistance to tear of SLC compared with both eyes. Ophthalmology 2016;123:265–74. 8 Carifi ,G Miller MH, Pitsas C, et al. Complications and outcomes of CCC and FLACS can be accounted by several factors including: phacoemulsification cataract surgery complicated by anterior capsule tear. Am J (1) doubling of the capsular edge thickness; (2) the rolled-over Ophthalmol 2015;159:463–9. folded edge which itself increases mechanical strength; (3) a 9 Marques FF, Marques DMV, Osher RH, et al. Fate of anterior capsule tears during smooth edge devoid of irregularity and defect, unlike FLACS cataract surgery. J Cataract Refract Surg 2006;32:1638–42. 10 Daya SM, Chee S-P, Ti S-E, et al. Parameters affecting anterior capsulotomy tear in particular which had a series of micro-undulations (postage strength and distension. J Cataract Refract Surg 2019;45:355–60. stamp appearance) and adjacent aberrant holes from pulses asso- 11 Werner L, Jia G, Sussman G, et al. Mechanized model to assess capsulorhexis ciated with multiple passes; both of which could promote frac- resistance to tearing. J Cataract Refract Surg 2010;36:1954–9. ture reducing resistance to tear; and (4) thermal change of the 12 J.H. M. Handbook of Biological Statistics. 3rd ed. Baltimore Maryland: Sparky House anterior capsule changing regular ordered Collagen Type IV to Publishing, 2014. 13 Unal M, Yücel I, Sarici A, et al. Phacoemulsification with topical anesthesia: resident amorphous collagen with increased elasticity. The latter permits experience. J Cataract Refract Surg 2006;32:1361–5. increased distension at a lower load before reaching the point 14 Olali CA, Ahmed S, Gupta M. Surgical outcome following breach rhexis. Eur J where the capsule stretches almost maximally and in a non-linear Ophthalmol 2007;17:565–70. manner ultimately leading to a tear. 15 Chan T, Pattamatta U, Butlin M, et al. Intereye comparison of femtosecond laser- assisted cataract surgery capsulotomy and manual capsulorhexis edge strength. J Hydrodissection during phacoemulsification can lead to Cataract Refract Surg 2017;43:480–5. prolapse of a lens nucleus during surgery potentially leading to 16 Auffarth GU, Reddy KP, Ritter R, et al. Comparison of the maximum applicable stretch an anterior capsule tear.8 9 The distension of the capsulotomy force after femtosecond laser-assisted and manual anterior capsulotomy. J Cataract perimeter for SLC in this study was found to average 28 mm Refract Surg 2013;39:105–9. 17 Friedman NJ, Palanker DV, Schuele G, et al. Femtosecond laser capsulotomy. J Cataract compared with 24 mm for CCC and 23 mm for FLACS. For a Refract Surg 2011;37:1189–98. 5.0 mm capsulotomy SLC would stretch 78%–85% to 8.9–9.25 18 Packer M, Teuma EV, Glasser A, et al. Defining the ideal femtosecond laser http://bjo.bmj.com/ mm and for both CCC and FLACS would stretch 46%–59% to capsulotomy. Br J Ophthalmol 2015;99:1137–42. 7.3–7.95 mm (table 3). This would theoretically result in less 19 Sándor GL, Kiss Z, Bocskai ZI, et al. Comparison of the mechanical properties of the anterior lens capsule following manual capsulorhexis and femtosecond laser chance of an anterior capsule tear during lens prolapse and capsulotomy. J Refract Surg 2014;30:660–4. supported by the pairwise analysis of perimeters resulting in SLC 20 Schultz T, Joachim SC, Noristani R, et al. Greater vertical spot spacing to improve being superior to, and more resistant to anterior tears than, CCC femtosecond laser capsulotomy quality. J Cataract Refract Surg 2017;43:353–7. and FLACS (p<0.01). 21 Sándor GL, Kiss Z, Bocskai ZI, et al. Evaluation of the mechanical properties of the anterior lens capsule following femtosecond laser capsulotomy at different pulse on 5 April 2019 by Andrew Webb. Protected copyright. In conclusion, this pairwise study of human cadaver eyes energy settings. J Refract Surg 2015;31:153–7. controlled for deviations in circularity and dimensional size, 22 Williams GP, George BL, Wong YR, et al. The effects of a low-energy, high frequency comparing three methods of capsulotomy demonstrated SLC liquid optic interface femtosecond laser system on lens capsulotomy. Sci Rep 2016;6. was significantly stronger than curvilinear capsulorhexis which 23 Barraquer RI, Michael R, Abreu R, et al. Human lens capsule thickness as a function of age and location along the sagittal lens perimeter. Invest Ophthalmol Vis Sci in turn was significantly stronger than femtosecond laser 2006;47:2053–60. capsulotomy. SEM findings demonstrated structural changes 24 Dick HB, Aliyeva SE, Hengerer F. Effect of trypan blue on the elasticity of the human supporting these findings. Clinically SLC may provide increased anterior lens capsule. J Cataract Refract Surg 2008;34:1367–73. surgical safety by reducing the chances of anterior capsular tears 25 Jaber R, Werner L, Fuller S, et al. Comparison of capsulorhexis resistance to tearing during cataract surgery. with and without trypan blue dye using a mechanized tensile strength model. J Cataract Refract Surg 2012;38:507–12. 26 Abell RG, Davies PEJ, Phelan D, et al. Anterior capsulotomy integrity after femtosecond Contributors SD wrote the manuscript. DHM was involved in study design, data laser-assisted cataract surgery. Ophthalmology 2014;121:17–24. and statistic analysis. S-PC and S-ET performed the experiments. RP made written 27 Al Harthi K, Al Shahwan S, Al Towerki A, et al. Comparison of the anterior contributions to portions of the manuscript. capsulotomy edge created by manual capsulorhexis and 2 femtosecond Funding The research was partially funded by Excel-Lens and also National Medical laser platforms: scanning electron microscopy study. J Cataract Refract Surg Research Council Singapore. 2014;40:2106–12. 28 Mastropasqua L, Toto L, Calienno R, et al. Scanning electron microscopy evaluation of Competing interests SD is a consultant and medical monitor to Bausch and Lomb capsulorhexis in femtosecond laser-assisted cataract surgery. J Cataract Refract Surg manufacturer of the Victus laser used in this study. He is also an equity shareholder 2013;39:1581–6. in Excel-Lens, manufacturer of the CAPSULaser device studied. RP and DHM are 29 Ostovic M, Klaproth OK, Hengerer FH, et al. Light microscopy and scanning electron equity shareholders in Excel-Lens, manufacturer of the CAPSULaser device studied. microscopy analysis of rigid curved interface femtosecond laser-assisted and manual Dr S-PC is a speaker for Bausch and Lomb. anterior capsulotomy. J Cataract Refract Surg 2013;39:1587–92.

6 Daya S, et al. Br J Ophthalmol 2019;0:1–6. doi:10.1136/bjophthalmol-2018-313421