Meta-analysis of the efficacy and safety of femtosecond laser-assisted phacoemulsification combined with multifocal intraocular lens implantation

Author: Ge HM, Yan WL, Xu XZ, Jiang Q

Geographical coverage: Australia, Slovakia, the Philippines, South Korea and China

Sector: Cataract treatment

Sub-sector: Efficacy and safety

Equity focus: Not reported

Study population: Patients with cataract

Review type: Effectiveness review

Quantitative synthesis method: Meta-analysis

Qualitative synthesis method: Not applicable

Background: Cataract is a leading cause of global blindness, especially among older adults, with a prevalence exceeding 90 % in those over 80 years of age. Conventional phacoemulsification with intra-ocular lens implantation is considered the standard surgical approach owing to its simplicity, safety and low complication rate. As patients’ expectations of visual quality rise, however, cataract surgery has evolved towards refractive goals. Femtosecond-laser-assisted cataract surgery (FLACS), introduced clinically in 2009, enables precise anterior capsulotomy and nucleus fragmentation. When combined with multifocal intra-ocular lenses (MFIOLs), which provide good vision at multiple distances, FLACS offers the potential for superior, customised outcomes. Consequently, FLACS with MFIOL implantation is becoming increasingly popular among patients who demand high-quality post-operative vision.

Objective:

To evaluate systematically the efficacy and safety of FLACS combined with MFIOL implantation, compared with conventional phacoemulsification (CP) with MFIOL implantation, using meta-analysis.

Main findings:

Evidence base included Eleven studies (six randomised controlled trials and five retrospective cohort studies) encompassing 1,045 eyes. One study contributed two comparator groups. Research settings were Australia (1), Slovakia (1), the Philippines (2), South Korea (1) and China (6).

Visual acuity: No significant differences in uncorrected distance visual acuity (UDVA) were found at any follow-up point. At three months post-operatively, the pooled mean difference was –0.03 logMAR (95 % CI –0.06 to 0.00; P = 0.05). Uncorrected near visual acuity (UNVA) was similar at one day, one week, three months and one year; a marginal advantage for FLACS–MFIOL emerged at one month (mean difference = 0.03 logMAR; 95 % CI 0.00 to 0.06; P = 0.04).

Refractive accuracy: Post-operative spherical equivalent did not differ significantly (mean difference = 0.06 D; 95 % CI –0.02 to 0.15; P = 0.14). FLACS–MFIOL showed superior pseudophakic accommodative power for distance (mean difference = 0.57 D; 95 % CI 0.42 to 0.72) and near vision (mean difference = 1.32 D; 95 % CI 0.99 to 1.64; both P < 0.001).

Surgical metrics: Cumulative dissipated energy (CDE) was significantly lower with FLACS–MFIOL for grade II nuclei (mean difference = –1.94; 95 % CI –2.59 to –1.30) and grade III nuclei (mean difference = –3.81; 95 % CI –5.66 to –1.96; both P < 0.01).

Corneal health: Endothelial cell density (ECD) was better preserved in the FLACS–MFIOL group (mean difference = 111.75 cells mm⁻²; 95 % CI 86.27 to 137.23; P < 0.01).

Methodology:

Searches of PubMed, Embase and the Cochrane Library, together with Chinese databases (CNKI, VIP, SinoMed, WanFang), covered January 2010 to November 2020. Studies comparing CP–MFIOL (control) with FLACS–MFIOL (intervention) were eligible. Two reviewers independently screened titles and abstracts, extracted data and assessed quality—using the Cochrane RoB tool for RCTs and the MINORS checklist for non-randomised studies—resolving disagreements by discussion or third-party arbitration. Pooled estimates were calculated with fixed- or random-effects models according to heterogeneity (I² statistic); sensitivity analyses and funnel plots assessed robustness and publication bias.

Applicability / external validity:

The inclusion of both randomised and retrospective designs varied surgical expertise and different MFIOL models may limit the generalisability of pooled estimates. Moreover, restricting the search to English- and Chinese-language publications could introduce language bias. Real-world implementation, particularly in resource-limited settings, warrants further study.

Geographic focus:

Australia, Slovakia, the Philippines, South Korea and China; no geographical limits were applied.

Summary of quality assessment:

Confidence in the conclusions is medium. The search strategy was broad, inclusion criteria were explicit, and dual-reviewer processes followed recognised standards. Study characteristics and risk-of-bias assessments were reported, and heterogeneity explored appropriately. Limitations include language restriction, absence of an excluded-studies list and insufficient stratification of findings by risk-of-bias status.