Changes in endothelial cell density after conventional phacoemulsification and FLACS in eyes with dense cataracts

FLACS with customized nuclear fragmentation patterns resulted in significantly less intraoperative phacoemulsification time, surgical time, and endothelial cell loss compared with conventional phacoemulsification for dense cataracts.

Abstract

Purpose:

To compare the safety, effectiveness, and changes in endothelial cell density (ECD) after standard phacoemulsification and femtosecond laser-assisted cataract surgery (FLACS) in dense cataracts (Lens Opacities Classification System III grade 3.0 NO and above).

Setting:

Nethradhama Superspeciality Eye Hospital, Bangalore, India.

Design:

Prospective, randomized, comparative study.

Method:

100 eyes of 100 patients were randomly assigned to either conventional phacoemulsification surgery (CPS) or FLACS (LENSAR Laser System) using predefined patterns of nuclear fragmentations. Measured outcomes included intraoperative phacoemulsification time, irrigation fluid volume, surgical time, and complication rates. ECD was evaluated at baseline, 2 weeks, and 6 months postoperatively.

Results:

The CPS group had significantly higher mean total phacoemulsification time (11.17 ± 6.5 seconds vs 8.03 ± 3.77 seconds), effective phacoemulsification time (6.14 ± 3.62 seconds vs 4.42 ± 2.07 seconds), and total surgical time (8.18 ± 2.36 minutes vs 7.11 ± 1.55 minutes) than the FLACS group (P < .001). Volume of irrigation fluid was comparable between the 2 groups (CPS, 36.7 ± 12.18 mL; FLACS, 38.64 ± 13.73 mL (P = .45). Mean corneal edema score and central corneal thickness values on postoperative day 1 were significantly higher for the CPS group. At 6 months, % ECD loss was significantly higher in the CPS group (16.08%) vs the FLACS group (12.8%) (P < .001). 3 eyes in the CPS group had wound burns of varying severity.

Conclusions:

FLACS with customized nuclear fragmentation patterns resulted in significantly less intraoperative phacoemulsification time, surgical time, and endothelial cell loss compared with conventional phacoemulsification when treating dense cataracts.

Trial Registration Number:

CTRI/2021/12/038887.

Dense cataracts predispose the eye to various complications during phacoemulsification such as loss of endothelial cell density (ECD), corneal edema, posterior capsular rupture, zonular dialysis, nucleus drop, and vitreous loss, which can potentially lead to corneal decompensation.¹ Several surgical approaches may be considered for the removal of dense cataracts including extracapsular cataract extraction (ECCE), small-incision extracapsular cataract extraction, phacoemulsification, and femtosecond laser-assisted cataract surgery (FLACS), depending on the skill of the surgeon and availability of technology.^2,3 Several studies have evaluated the safety and effectiveness of FLACS, demonstrating beneficial effects such as reduction in effective phacoemulsification time (EPT), endothelial cell loss, and postoperative corneal edema, which are associated with improved visual outcomes and more rapid postoperative recovery.^4–10

The LENSAR Laser System (LENSAR, Inc.) incorporates Augmented Reality through advanced Scheimpflug imaging, ray tracing, and 3D confocal structured illumination (3D-CSI) to construct a 3D model of the anterior segment of the eye including the cornea and lens.¹¹ The software consolidates 16 individual high-resolution Scheimpflug images from 8 angular positions to identify the major interfaces, that is, anterior and posterior corneal surface and anterior and posterior lens capsule, details of the location and density of the lens nucleus, and the boundaries between endonucleus, epinucleus, and cortex, thus allowing automated grading of cataract density and customization of phacofragmentation, aiming at reducing cumulative dissipated energy (CDE) and minimizing potential for inflammation and endothelial cell loss.

This prospective, randomized study was conducted to investigate whether automated cataract density grading and customized fragmentation patterns in the LENSAR Laser System offer any significant advantages for intraoperative ultrasound time, volume of fluid consumed, and ECD in dense cataracts when compared with standard phacoemulsification surgery over a period of 6-month postoperative follow-up.

METHODS

This prospective, interventional, randomized, comparative study was approved by the institutional ethics committee of Nethradhama Super Speciality Hospital, Bangalore, and adhered to the tenets of Declaration of Helsinki. Verbal and written informed consent was obtained from all patients participating in the study. The study included 100 eyes from 100 patients fulfilling the eligibility criteria. Patients were assigned to receive either conventional phacoemulsification surgery (CPS) or FLACS using computer-generated random numbers.

The primary end point evaluated was the change in ECD from baseline to 6 months postoperative. Secondary end points included total phacoemulsification time (TPT) and effective phacoemulsification time (EPT), volume of irrigation fluid, total surgical time (incision to closure), total time for phacoemulsification and irrigation/aspiration (I/A), and incidence of intraoperative complications.

Patients of 40 to 80 years with cataract of Lens Opacities Classification System (LOCS) III grade 3.0 or more in the study eye willing and able to understand and sign the informed consent, and able to attend the postoperative examinations per protocol schedule were included. Patients with baseline ECD less than 1500 cells/mm²; pupillary diameter less than 6 mm; complex cataract, such as subluxated cataract, post uveitis cataract, or traumatic cataract; leucomatous corneal opacity; grade 2 or greater pterygium; pregnant or lactating women; and immunocompromised patients were excluded.

Detailed preoperative evaluation included visual acuity assessment with the ETDRS chart at 4 m, slitlamp examination for anterior segment evaluation and cataract grading, dilated fundus examination for posterior segment evaluation, intraocular pressure recording by noncontact tonometry (Nidek Noncontact Tonometer NT-510), specular microscopy for ECD (Tomey Specular microscope EM-4000), and optical biometry using IOL Master 700 (Carl Zeiss Meditec AG).

All surgeries in both groups were performed by a single experienced surgeon (S.G.) under topical anesthesia with proparacaine and intracameral 2% preservative-free lignocaine. The phacoemulsification machine used for cataract extraction in both the groups was the CataRhex 3 phacoemulsification system (Oertli Instruments AG).

For the FLACS group, the LENSAR femtosecond laser was used to create a 5.2 mm anterior capsulotomy. Customized nuclear fragmentation patterns were automatically selected by the laser based on the nuclear density (Figure 1). The augmented reality feature grades the nucleus into 1 of the 5 categories, and on selecting the “custom frag” feature, it applies 1 of the 5 preloaded fragmentation patterns, most suited for the type of cataract (Figure 2). After the laser delivery, the patient was shifted to the operating bed and brought under the operating microscope for extraction of the prechopped nuclear fragments.

Figure 1.

Augmented reality enabled nucleus grading by the LENSAR FS laser system

Figure 2.

Morphology of custom fragmentation pattern templates 1, 2, 3, 4, and 5 available in the LENSAR FS laser system used in the study

For the CPS group, standard phacoemulsification was performed with a 2.8 mm incision. In both groups, after hydrodissection and hydrodelineation, the nucleus was emulsified in longitudinal phacoemulsification mode using hard cataract settings, wherein an ultrasound power of 55% to 60% in pulse mode, a flow rate of 45 mL/min, and a vacuum of 540 mm Hg were used.

A dispersive OVD (Viscoat, Alcon Laboratories, Inc.) was used every 60 seconds during phacoemulsification in both the groups. After phacoemulsification, the cortex was removed using a coaxial I/A handpiece and the intraocular lens was implanted into the capsular bag under balanced salt solution. At the end of surgery, corneal incisions were sealed by stromal hydration.

At the end of each surgical procedure, TPT (seconds), EPT (seconds), fluid volume/case (cc), total surgical time (incision to closure), and combined phaco and I/A time (minutes) were noted. For measurement of fluid used per case, the priming fluid was first discarded from the drainage bag before the start of each case. Immediately after the surgery, the drainage bag was removed from the consumable assembly and the aspirate was carefully emptied into a measuring beaker. The upper fluid level was noted by a single observer, who was masked to the patient's group assignment.

Postoperative examinations were performed at 1 day, 2 weeks, and 6 months, and included the assessment of uncorrected distance visual acuity (UDVA), corrected distance visual acuity (CDVA), noncontact tonometry, slitlamp examination for corneal edema grading, and specular microscopy for central corneal thickness and ECD.¹²

Statistical Analysis

At least 110 eyes were required to prove that the null hypothesis mean change in ECD for the 2 groups, that is, FLACS and CPS, is same vs the alternative hypothesis of unequal mean change in ECD between the groups. To achieve 90% power with 5% level of significance and considering 10% drop-out rate, 100 participants were required to enrol. SPSS software for Windows v. 17.0.0 (IBM Corp.) was used for statistical analysis. All values were expressed as mean ± SD. Data were checked for normality before subjecting to analysis. The independent sample t test was used for intergroup comparison, and the paired t test was used for intragroup comparison of means. A P value of 0.05 or less was considered statistically significant.

RESULTS

A total of 100 eyes from 100 eligible patients were included in the study. Supplemental file 1 (available at http://links.lww.com/JRS/B69) gives the preoperative characteristics and demographic data of patients included in the study. The 2 groups were comparable about the age, K1, K2, axial length, anterior chamber depth, lens thickness, white-to-white diameter, central corneal thickness, and ECD.

Intraoperative Parameters

Intraoperatively, the TPT was significantly higher in the CPS group (11.17 ± 6.5 seconds) vs the FLACS group (8.03 ± 3.77 seconds) (P < .001). Similarly, the EPT was significantly higher in the CPS group (6.14 ± 3.62 seconds) when compared with the FLACS group (4.42 ± 2.07 seconds) (P < .001). Regarding the mean surgical time and phaco + I/A time, these were significantly higher in the CPS group (8.18 ± 2.36 minutes and 5.98 ± 1.80 minutes) when compared with the FLACS group (7.11 ± 1.55 minutes and 4.58 ± 1.95 minutes) (P < .00). The fluid volume per case in the CPS group was 36.7 ± 12.18 mL, which was comparable with the FLACS group (38.64 ± 13.73 mL) (P = .45, Table 1).

Table 1.

Comparison of intraoperative parameters of both the study groups

Parameter	Conventional phacoemulsification	LENSAR	P value
IOL power (D)	20.96 ± 2.69	21.07 ± 2.27	.82
Grade of cataract	3.62 ± 0.6	3.88 ± 0.7	.06
Phaco time (s)	11.17 ± 6.5	8.03 ± 3.77	<.001
Effective phaco time (s)	6.14 ± 3.62	4.42 ± 2.07	<.001
Fluid (mL)	36.7 ± 12.18	38.64 ± 13.73	.45
Total surgical time (min)	8.18 ± 2.36	7.11 ± 1.55	<.001
Phaco + I/A time (min)	5.98 ± 1.80	4.58 ± 1.95	<.001

I/A = irrigation/aspiration

Visual Acuity

At 6 months, 58% (n = 29) of eyes in the CPS group and 68% (n = 34) of eyes in the FLACS group had cumulative UDVA of 20/20 or better. One hundred percent of eyes had a cumulative CDVA of 20/25 or better in the CPS group vs 100% 20/20 or better in the FLACS group (Supplemental file 3, available at http://links.lww.com/JRS/B71). No significant differences were observed for mean UDVA and CDVA between the 2 groups at 6 months (P = .17 and .15, respectively, Table 2).

Table 2.

Comparison of postoperative outcomes of both the study groups at POD-1, 2 weeks and 6 months

Parameter	Conventional phacoemulsification (CPS)	FLACS (LENSAR)	P value
POD-1
Corneal clarity grade	0.90 ± 0.93	0.50 ± 0.83	.01
CCT	566 ± 30.1	542 ± 27.88	<.001
2 wk postop
UDVA (logMAR)	0.15 ± 0.14	0.10 ± 0.12	.03
CDVA (logMAR)	0.01 ± 0.06	0.004 ± 0.03	.52
Sphere (D)	0.07 ± 0.4	0.02 ± 0.4	.55
Cylinder (D)	−0.38 ± 0.6	−0.2 ± 0.48	.05
SE (D)	−0.11 ± 0.4	−0.08 ± 0.48	.7
CCT	540 ± 32.16	536 ± 29.75	.2
6 mo postop
UDVA (logMAR)	0.07 ± 0.09	0.05 ± 0.08	.17
CDVA (logMAR)	0 ± 0	−0.004 ± 0.01	.15
Sphere (D)	0.15 ± 0.3	−0.01 ± 0.15	.00
Cylinder (D)	−0.19 ± 0.4	−0.07 ± 0.33	.13
SE (D)	0.05 ± 0.36	−0.04 ± 0.23	.10
CCT	536 ± 25.40	532 ± 28.21	.20

CCT = central corneal thickness; CPS = conventional phacoemulsification surgery; POD = postoperative day; SE = spherical equivalent

Refraction

Spherical equivalent (SE) refractive accuracy within ±0.50 diopter (D) was achieved in 94% (n = 47) of eyes in the CPS group vs 98% (n = 49) of eyes in the FLACS group at 6 months. All eyes in both groups were within ±1.00 D SE (Supplemental file 4, available at http://links.lww.com/JRS/B72). The mean SE in the CPS group at 6 months was 0.05 ± 0.36 D vs −0.04 ± 0.23 D, which was statistically insignificant between groups (P = .10).

Endothelial Cell Density

Supplemental file 5 (available at http://links.lww.com/JRS/B73) shows the distribution of the various custom fragmentation patterns used in the FLACS group of the study, and Table 3 summarizes the energy and spot spacing parameters used for each of these fragmentation patterns.

Table 3.

Energy and spot spacing parameters used for each of the custom fragmentation patterns

Parameter	Category 1	Category 2	Category 3	Category 4	Category 5
Energy (μJ)	10	10	10	10	14.0
PRF (kHz)	80	80	80	80	20
Thick spacing XY (μm)	10	10	25	25	10
Line spacing Z (μm)	20	20	20	20	25
Shot spacing (μm)	10	10	10	10	20
Max turret radius (mm)	0.00	0.00	1.00	1.00	2.00
Extend turret (mm)	−0.50	−0.50	−0.50	−0.50	-

PRF = pulse repetition frequency

At 2 weeks and 6 months postoperatively, the mean ECD values were significantly higher in the FLACS group compared with the CPS group (P-values <0.001 for both timepoints). At 2 weeks, the percentage of ECD loss was higher (15.4%) in the CPS group compared with the FLACS group (12.4%) (P = .02). Similarly, at 6 months, the percentage of ECD loss was significantly higher in the CPS group (16.08%) vs the FLACS group (12.8%) (P < .001, Table 4).

Table 4.

Postoperative endothelial cell loss at 2 weeks and 6 months postoperatively

Parameter	Pre-ECD (cells/mm2)	2 wk ECD (cells/mm2)	ECD loss cells (%)	6 mo ECD (cells/mm2)	ECD loss cells (%)
Conventional phacoemulsification	2540.5 ± 237.52	2148.42 ± 221.87	392 (15.4)	2132.4 ± 234.12	408 (16.08)
LENSAR	2523.62 ± 241.96	2210.16 ± 210.60	313 (12.4)	2198.2 ± 245.19	325 (12.8)
P value	.7	<.001		<.001

ECD = endothelial cell density

Complications

Postoperative corneal edema on postoperative day 1 showed a higher mean score for the CPS group (0.90 ± 0.93) compared with the FLACS group (0.50 ± 0.83) (P = .01). However, all corneas were clear by 2 weeks in both groups. The postop day 1 central corneal thickness (CCT) values showed increased thickness in both groups compared with baseline; however, the mean CCT values were significantly higher in the CPS group (566 ± 30.1 μm) vs the FLACS group (542 × 27.88 μm) (P < .001, Table 2). However, the CCT values were comparable at 2 weeks and 6 months.

Two eyes in the CPS group (LOCS grade 4) had mild wound burn not requiring intervention, whereas another eye (LOCS grade 5) had moderate grade wound burn that required suturing. There were no instances of ruptured capsule, cystoid macular edema, or retained lens fragments.

DISCUSSION

Various studies have compared the effects of FLACS and CPS on clinical outcomes and postoperative ECD (Supplemental file 2, available at http://links.lww.com/JRS/B70.^5,8,13–15 The mean ECD loss after CPS in these studies for various grades of cataracts at 3 to 6 months postop has been reported between 8% and 30%. The mean ECD loss after FLACS has ranged from 5% to 15.8%. However, the Cochrane review performed by Narayan et al. in 2023, included 42 randomized controlled trials in Europe, North America, South America, and Asia showed low-certainty evidence of little or no important difference in endothelial cell loss when comparing FLACS with CPS (MD 12 cells per mm² in favor of FLACS).¹⁶

Schroeter et al. compared ECD loss in cataracts between LOCS 1 to 3 for phacoemulsification operated by the Oertli Catarax system vs the Femto LDV femtosecond laser system and showed that the phacoemulsification group had significantly higher ECD loss of 29.6% vs 15.8% in the FLACS group, 3 months postoperatively.¹⁷ A study by Kelker et al. evaluated ECD loss after phacoemulsification using the Stellaris system in LOCS grades 1 to 3 and found an average loss of 281 cells/mm² (13.7%) vs 134 cells/mm² (8%) in a comparative group using the Catalys FLACS system at 6 months postoperatively.¹⁸

Mencucci et al. compared ECD loss in LOCS 2 to 3 grade nuclei using the Centurion phacoemulsification system with a comparative group of FLACS using the LenSx femtosecond system and observed a higher ECD loss of 8.23% 6 months postoperatively in the phacoemulsification group vs 5.85% in the FLACS group.¹⁹ Krarup et al. conducted a similar study for LOCS grade 2 cataracts, wherein phacoemulsification was accomplished using the Centurion phacoemulsification system and FLACS using the LENSAR Laser System.²⁰ Results showed that the mean ECD loss in the phacoemulsification group was 465 cells/mm (17%) vs 365 cells/mm (13.5%) in the FLACS group, 6 months postoperatively.

Few authors have studied grades of cataracts above LOCS III. Conrad-Hengerer et al. observed a 13.7% ECD loss in the phacoemulsification group vs 8% ECD loss with FLACS with LOCS grades of 1 to 4.⁸ The phacoemulsification system used in this study was the Stellaris PC system, and the FLACS system used was the Catalys. Vasavada et al. also evaluated ECD loss in LOCS grades 1 to 4 and compared the same between conventional phacoemulsification (Centurion) and FLACS (LenSx) and found that ECD loss was 9.20% in the phacoemulsification group, when compared with 8.25% in the FLACS group.²¹ On the other hand, Daliya et al. also conducted a very similar study on LOCS grade 1 to 4 nuclei, using the same systems for phacoemulsification and FLACS, and observed an average ECD loss of 11.2% in the phacoemulsification group when compared with the FLACS group (8%).²² These results, in general, favor the potential beneficial effects of prenucleus fragmentation and reduced need for ultrasound energy accomplished by FLACS. Yesilirmak et al. conducted studies in 2016 and 2018, comparing CDE expenditure in various groups.^14,15 In 2016 study, the CDE expenditure with 2 phacoemulsification platforms the gravity fluidics torsional Infiniti and the active fluidics torsional Centurion after conventional cataract surgery or FLACS (Lensx) was studied. It showed that patients who had pretreatment with the femtosecond laser (Lensx) had significantly lower CDE when the active fluidics platform (5.18 ± 4.58) was used for phacoemulsification than when the gravity fluidics platform (7.00 ± 6.85) was used.¹⁴ The 2018 study, however, investigated the differences in NRT and CDE between traditional cataract surgery and 3 different femtosecond laser platforms and analyzed how these parameters were affected by surgeon experience. There was no significant CDE difference (P = .05) with each FLACS platform when compared with traditional surgery in both the attending and fellow groups for grade 2 to 3 cataracts.¹⁵

This study is unique in the sense that we included only denser grades of cataracts, that is, LOCS grade 3 and above, which are more challenging surgically and result in greater endothelial cell loss when compared with softer grades of cataracts. The percentage of ECD loss was not only significantly less in the FLACS group (12.8%) compared with the CSP group (16%), and it was also fairly acceptable when compared with the cumulative ECD loss range of 5% to 15.8% reported for FLACS in the literature (Supplemental file 2, available at http://links.lww.com/JRS/B70). The unique features of augmented reality, automated cataract grading, and use of customized fragmentation patterns may have played a favorable role in reducing the TPT and EPT, resulting in reduced endothelial cell loss, as the laser energy and spot spacing parameters selected in these patterns are optimized for the density of the nucleus. These optimized patterns potentially result in more efficient fragmentation of the nucleus, because the nucleus grading is automated compared with subjective grading of the nucleus by the surgeon, and use of a standard laser settings of “soft” or “hard” cataract created by the surgeon. The customized patterns address 5 densities of nuclei and thus automatically select 1 of the 5 preloaded patterns, most suited for and matching the type of cataract, thus making the fragmentation much more specific to the grade of the nucleus.

Regarding the visual and refractive outcomes, no significant differences were observed for the mean UDVA, CDVA, and SE accuracy between the 2 study groups at 6 months, suggesting that CPS was safe and effective and delivered comparable results as FLACS. However, the significantly high CCT values on postop day 1 in the CSP group would suggest greater corneal edema, due to a combination of significantly higher TPT, EPT, and overall surgical time in this group, when compared with the FLACS group.

In conclusion, FLACS with customized nuclear fragmentation patterns resulted in significantly less intraoperative phacoemulsification time, surgical time, and endothelial cell loss compared with conventional phacoemulsification while treating denser grade cataracts. Further data and studies evaluating the potential advantages of automated nucleus grading and use of customized laser patterns vs noncustomized/standard patterns are suggested to ascertain the perceivable advantages of this technology.

WHAT WAS KNOWN

• FLACS has demonstrated beneficial effects vs standard conventional phacoemulsification surgery (CPS) such as significant reduction in effective phacoemulsification time, reduced endothelial cell loss, and postoperative corneal edema, which were associated with improved visual outcomes and postoperative recovery.

• Reported range of cumulative endothelial cell density (ECD) loss (5% to 15.8%) involved studies with variable degrees of LOCS densities mostly including LOCS grades 1 to 3 and some including LOCS grades 1 to 4.

WHAT THIS PAPER ADDS

• In our study, involving exclusively the denser grades of cataracts, that is, LOCS grade 3 and above, the average ECD loss at 6 months was 12.8%, suggesting a favorable outcome for endothelial health in the long term.

• The unique feature of augmented reality, automated cataract grading, and use of customized fragmentation patterns potentially played a beneficial role in reducing the total and effective phacoemulsification time and reduced endothelial cell loss, as the laser energy and spot spacing parameters selected in these patterns were optimized for the density of the nucleus.