This retrospective case series included patients who had previously undergone LASIK or PRK for myopia and underwent femtosecond laser assisted phacoemulsification and trifocal IOL (AT LISA tri 839MP, Carl Zeiss Meditec AG, Jena, Germany) implantation in AIER Group’s Eye Hospitals (Beijing Aier-Intech Eye Hospital, Guangzhou Aier Eye Hospital, Chongqing Aier Eye Hospital, Wanzhou Aier Eye Hospital and Shenzhen Aier Eye Hospital) from January 2017 to May 2019. Ethics Committee approval by the Institutional Review Board of the Aier School of Ophthalmology of Central South University was obtained for the present study protocol that adhered to the tenets of the Declaration of Helsinki.
The inclusion criteria consisted of eyes (1) that had surgery indication for cataract surgery without any contraindications of ocular surgical therapy in the preoperative examination, (2) with corneal astigmatism ≤ 1.5 D and the angles of Kappa and Alpha both being < 0.3 mm, and (3) had no complications of posterior capsular rupture or zonular dialysis during the cataract surgery. The exclusion criteria included eyes that (1) had decentered ablation (decentration > 0.5 mm), corneal scar, retinoschisis, haze, myopic retinopathy, retinoschisis or retinal detachment after myopia excimer laser correction, (2) with corneal astigmatism ≥ 1.5 D or irregular astigmatism ≥ 0.3 µm, and (3) had inflammation, glaucoma or other diseases that might affect the multifocal IOL implantation.
Enrolled patients all underwent preoperative ophthalmologic examinations including corrected distance visual acuity (CDVA), UDVA, manifest refraction, retinal visual acuity, intraocular pressure, ocular A and B-scan ultrasonography, non-contact specular microscope, optical coherence tomography (OCT) and ray tracing aberrometry. The retinal visual acuity was assessed with Lambda 100 retinometer (Heine, Germany). The axial length, keratometry and anterior chamber depth were measured using the LenStar LS900 (Haag Streit, Switzerland).
The postoperative measurements at 1-day, 1-month and 3-month included CDVA, UDVA, UIVA, UNVA, and the subjective manifest refractions (spherical equivalent) as well as their changes. Through-focus monocular logMAR acuity (defocus curve) was also measured. The spectacle independence rate, satisfaction, and visual symptoms were recorded at 3-month postoperatively with the Visual Function index 14 (VF-14) questionnaire, which provides an index of functional impairment in patients with cataract [21, 22], as well as another short questionnaire concerning negative visual symptoms. VF-14 was translated into Chinese  and based on 14 uncorrected vision–dependent daily activities, scoring each item with regard to the degree of difficulty as follows: no difficulty (4 score), a little difficulty (3 score), moderate difficulty (2 score), quite difficult (1 score), or impossible to perform the task (0 score). Items were not included in the scoring if patients could not perform the activity for reasons other than vision-related. Scores on all activities were averaged, and the mean score was then multiplied by 25. The resulting VF-14 score ranged from 0 (worst functional impairment) to 100 (no disability) . The short questionnaire with regard to some negative visual symptoms commonly observed after cataract surgery, such as halo, glare, was recorded by the correspondent surgeon.
All the surgeries were performed combining femtosecond laser-assisted phacoemulsification and intraocular lens implantation by experienced surgeons. For the eyes with corneal astigmatism between 0.75 D and 1.5 D (6 eyes), femtosecond laser-assisted corneal relaxing incision was performed for correction. Preoperatively, the surgeons used tropicamide to maintain pupil dilation intraoperatively. Under topical anaesthesia, capsulotomies (diameters were all set as 5.2 mm), lens fragmentation and corneal relaxing incisions were performed subsequently using the LenSx femtosecond laser (Alcon Laboratories, Inc, Fort Worth, Texas, USA). Phacoemulsification was then performed using standard ultrasound technique. In case of the patients with high myopia who underwent femtosecond laser surgery, both anterior and posterior capsules were thoroughly polished to reduce the risk of capsule contraction. The trifocal IOL (AT LISA tri 839MP) was implanted in the capsular bag using an injector. The residual ophthalmic viscosurgical device was removed, and the position of the lens was adjusted. All incisions were hydrated and the patients’ conjunctival sac was treated with dexamethasone-tobramycin ophthalmic ointment.
For lens power calculation, a multi-formula average method was performed, in which 4 formulas of Hagis-L , Barrett True K , Shammas No-History  and ray-tracing methods  setting the target refraction as postoperative emmetropia were used. The implanted IOL power was the average of the calculated results from all 4 formulas. For the first treated eye of patients who underwent bilateral implantation and the eyes with a large difference (> 1.0 D) in the IOL power calculations using the 4 formulas, a modified aphakic refraction technique was applied based on the reported aphakic refraction procedure by Dr. Mackool . At first, the cataract removal was performed without IOL implantation. 1 Day later, manifest refraction examination was performed for the lens power calculation followed by the IOL insertion. All the two-staged-procedure surgeries were performed in the same fashion after obtaining patient’s consent. After one week, another eye surgery would be performed.
Postoperative treatment included one drop of tobramycin-dexamethasone eye drops every two hours for 3 days, afterward four times per day until 2 weeks, and then three times per day for another 2 weeks; and one drop of levofloxacin eye drops four times every day for 1 week; pranoprofen and sodium hyaluronate eye drops were administered as appropriate.
Statistical analysis was performed using Microsoft Excel. Mean (± SD) was reported for continuous variables. Normal probability plots and Kolmogorov-Smirnov and Shapiro-Wilk tests were used to check the normality of data in SPSS software. Pre- and post-UDVA, pre- and post-CDVA outcomes were compared with the t test. Nonparametric tests were used to evaluate differences within groups. Differences were considered statistically significant when the P value was less than 0.05.