FormalPara Key Summary Points

Why carry out this study?

The question of whether femtosecond laser-assisted cataract surgery (FLACS) has an effect on the fundus has been a focus of clinical research.

From this perspective, we conducted a short-term assessment of the changes in central macular thickness (CMT) and subfoveal choroidal thickness (SFCT) in diabetic patients after phacoemulsification (PE) surgery and FLACS combined with intraocular lens (IOL) implantation.

What was learned from the study?

Both FLACS and traditional PE surgery caused an increase in CMT and SFCT at week 1 to month 3 after surgery.

The CMT and SFCT in the FLACS group recovered to the preoperative level earlier than the PE group.

Baseline CMT was an important predictor of postoperative CMT changes in both groups.

The BCVA in the FLACS group was batter than that in PE group at week 1 to month 3 after surgery.

Digital Features

This article is published with digital features, including a summary slide, to facilitate understanding of the article. To view digital features for this article go to


According to data published by the International Diabetes Federation, the number of adults with diabetes worldwide was predicted to reach approximately 463 million in 2019, and was expected to increase to 700 million by 2045. Diabetes mellitus (DM) can destroy the blood–retinal barrier [1, 2] and cause pathological changes in the retina [3], known as diabetic retinopathy (DR). Data from 35 studies in the United States, Australia, Europe and Asia show that the overall prevalence of DR in diabetic patients is 34.6%, and the prevalence of proliferative diabetic retinopathy (PDR) is 7.0% [4].

In addition, diabetes mellitus increases the risk of cataracts, especially at younger ages, and the turbidity of the lens hinders visualization of the patient’s retina during funduscopy [5, 6]. The presence of cataract was also found to be the most significant risk factor for visual impairment in diabetic patients [7, 8]. It is estimated that up to 20% of cataract surgeries are performed in patients with diabetes mellitus [9,10,11]. Therefore, cataract surgery is essential to restore vision in these patients and to diagnose and monitor the development of DR.

Phacoemulsification (PE) surgery can cause an increase in inflammatory factors such as IL-6 and prostaglandin in the anterior chamber and vitreous body. Simultaneously, the heat, radiation and shock generated by PE have a greater influence on the vitreous and fundus [12, 13]. After the operation, some diabetic patients have an accelerated course of fundus microangiopathy [14], retinal thickening [15] and choroidal thickening, with different changes shown among studies [16, 17]. Researchers have confirmed that femtosecond laser-assisted cataract surgery (FLACS) can optimize PE parameters [18], reduce the cumulative release of ultrasound energy [19] and reduce complications [20, 21]. However, it is worth noting that FLACS requires suction of the cornea, and so the intraocular pressure is increased significantly during capsulotomy and fragmentation [22]. Some scholars believe that FLACS causes more serious inflammation and increases the incidence of complications [23].

Therefore, whether FLACS has an effect on the fundus has been the focus of clinical research. Especially for diabetic patients with strong postoperative inflammatory response and who are prone to macular edema, whether FLACS is safe and effective requires further research evidence. From this perspective, we conducted a short-term assessment of the changes in central macular thickness (CMT) and subfoveal choroidal thickness (SFCT) of diabetic patients after PE surgery and FLACS combined with intraocular lens (IOL) implantation.



This prospective study recruited 91 diabetic patients (91 eyes) undergoing PE surgery or FLACS combined with IOL implantation in Shanghai Xin Shi Jie Eye Hospital (China). The study complied with the Declaration of Helsinki of 1964 and its later amendments. The study was reviewed and approved by the Shanghai Xin Shi Jie Eye Hospital ethics committee. All patients understood and provided written informed consent. All patients had been diagnosed with diabetes mellitus and were currently being treated with orally administered drugs or insulin injections. Fasting blood glucose was controlled at 4.4–10 mmol/L and HbA1c (%) was less than 8% before surgery. Patients were diagnosed as having no diabetic retinopathy (NoDR) or mild non-proliferative diabetic retinopathy (NPDR) according to the International Clinical Disease Severity Scale for DR. The first eye was enrolled if the patient underwent surgery for both eyes. Patients with DR of other stages, macular edema, other ocular diseases or history of ocular surgery, long-term local or systemic application of steroid drugs, preoperative lens opacity obviously making it difficult to obtain optical coherence tomography (OCT) scan images, diopter >  −6.00 D or axial length > 26.0 mm, and patients with severe hypertension, smoking or alcoholism were excluded. The patients were divided into two groups according to patient preference; the PE group underwent PE surgery combined with IOL implantation, and the FLACS group underwent FLACS combined with IOL implantation.

Patient Assessment

Best-corrected visual acuity (BCVA) was examined using the Snellen decimal scale and then converted to the logarithm of the minimal angle of resolution (logMAR) scale. Intraocular pressure was measured using a noncontact technique (FT-1000 tonometer; TOMEY Corporation, Nagoya, Japan). In addition, all patients underwent slit-lamp examination (SL-3G, Topcon Corporation, Tokyo, Japan), diopter (CT-80, Topcon), axial length (IOLMaster 500; Carl Zeiss Meditec Inc, Jena, Germany), wide-angle fundus photography (Daytona p200T, Optos plc, Dunfermline, UK), and macular and optic nerve examination before the operation to exclude other eye diseases. Based on the results of slit-lamp examination at maximum illumination without light filtering, the Lens Opacities Classification System (LOCS) III was used to grade the lens severity. BCVA, slit-lamp examination, intraocular pressure, wide-angle fundus photography and OCT examination were performed at 1 week and 1, 3, 6 and 12 months after surgery.


The same professional examiner used OCT (Cirrus HD-OCT 5000; Carl Zeiss Meditec, Inc., Dublin, CA, USA) to scan the macula with the macular cube 512 × 128 combo mode to automatically measure the CMT. SFCT was measured using the enhanced depth imaging (EDI) mode of the OCT (Spectralis HD-OCT; Heidelberg Engineering GmbH, Heidelberg, Germany). Under posterior segment tracking, 100 frames were captured and averaged to compose an A-scan; each EDI-OCT image had 768 A-scans. Using the sections through the center of the macula, the distance from the hyperreflective band corresponding to Bruch’s membrane under the foveola to the choroid–sclera junction was measured using the instrument’s software. The same experienced retinal expert carried out three measurements manually and took the average of the three. The examiner and the vitreoretinal doctor were unaware of the identity of patients at all times. All OCT examinations were conducted from 9 to 11 a.m. after pupil dilation.


All surgeries were performed by the same experienced surgeon. All patients were given 0.5% levofloxacin eye drops (Cravit; Santen Pharmaceutical Co., Japan) to prevent infections, and patients in the FLACS group were also given diclofenac sodium (Shenyang Sinqi, China) 3 days before surgery. Combination tropicamide and phenylephrine hydrochloride eye drops (Mydrin-P; Santen Pharmaceutical Co., Japan) were used for mydriasis 60 min before surgery, and 0.4% oxybuprocaine eye drops (Benoxil; Santen Pharmaceutical Co., Japan) were used for topical anesthesia.

In the FLACS group, the surgeon used the same LenSx Laser system (Alcon Laboratories Inc., Fort Worth, TX, USA) to perform the capsulotomy and fragmentation, with laser pulse energy of 5 mJ. Both groups used the same Infiniti system (Alcon Laboratories Inc., Fort Worth, TX, USA) for PE. After removing the cortex, the folded posterior chamber IOL (Rayner Intraocular Lenses Limited or HumanOptics AG) was implanted into the capsular bag, and the effective phaco time (EPT) was recorded. Postoperative drugs were 0.5% levofloxacin eye drops (Cravit Ophthalmic Solution; Santen Pharmaceutical Co., Ltd,, Japan) and tobramycin dexamethasone eye drops (Tobradex; S.A. Alcon-Couvreur N.V., Belgium), one drop four times a day for 2 weeks; the two groups were the same. Then the steroid drops were decreased to twice a day for a week.

Statistical Analysis

All data were statistically analyzed using SPSS 25.0 software (IBM Corp., Armonk, NY, USA). Mean ± standard deviation was used to describe the continuous variable of normal distribution, and non-normally distributed variables were expressed as median (interquartile range [IQR]). The t test was used to compare the normally distributed variables, the Wilcoxon test was used to compare the non-normally distributed variables, and the chi-square test was used to compare the measurement data. A multiple linear regression model was used to analyze the impact of the baseline parameters on the changes in CMT and CT versus baseline. P < 0.05 was considered to indicate a significant difference.



In total, 47 people (47 eyes) were followed up in the PE group, including 19 men and 28 women, and five patients were lost to follow-up. The average age of the patients in the PE group was 69.32 ± 6.24 (range 54–83) years. The FLACS group completed 44 follow-ups (44 eyes), including 26 men and 18 women, and three patients were lost to follow-up. The average age of the patients in the FLACS group was 70.77 ± 7.24 (range 53–85) years (Table 1).

Table 1 Characteristics of the PE and FLACS groups

No serious complications occurred during or after the operation. No patient developed pseudophakic cystoid macular edema (PCME) or subretinal fluid. There was no significant difference in age (P = 0.307), diabetic history (P = 0.454), HbA1c (%) (P = 0.544), anterior chamber depth (P = 0.792), lens grade (P = 0.827), ametropia (P = 0.243), axial length (P = 0.781), intraocular pressure (P = 0.118), BCVA (P = 1.000), CMT (P = 0.256) or SFCT (P = 0.318) between the two groups. The EPT in the PE group was 6.60 s (5.20–7.70 s), whereas it was 2.34 ± 0.80 s in the FLACS group, which was a significant decrease (P = 0.000) (Table 1).


The BCVA before and after surgery is shown in Table 2. A significant difference was observed between the two groups 1 week and 1 month after operation (P < 0.05). In both groups, the difference between the baseline and postoperative BCVA was statistically significant (P < 0.05) (Table 2).

Table 2 Preoperative and postoperative values of BCVA, CMT and SFCT

Preoperative and Postoperative CMT

The CMT before and after operation is shown in Table 2. No significant differences were found between the two groups at baseline or any postoperative follow-up points (P > 0.05). In the PE group, the difference between the baseline and postoperative CMT values was statistically significant at all time points (P < 0.05), and the same was true in the FLACS group except at month 12 (P = 0.859) (Tables 2, 3).

Table 3 Paired-samples test of postoperative CMT and SFCT compared with baseline

Preoperative and Postoperative SFCT

The SFCT values before and after operation are shown in Table 2. The difference in SFCT between the two groups was statistically significant only at postoperative month 12 (P = 0.033). A significant difference between the baseline and postoperative SFCT values was observed in the PE group (P < 0.05). Compared to the baseline SFCT, postoperative SFCT was significantly different in the FLACS group (P < 0.05) at week 1 and months 1 and 3. No significant difference in SFCT was observed between baseline and month 6 or 12 after operation in the FLACS group (P > 0.05) (Tables 2, 3; Fig. 1).

Fig. 1
figure 1

Trends of postoperative CMT and SFCT

Multiple Linear Regression Analysis of the PE Group

In the PE group, baseline CMT predicted postoperative CMT at week 1 and months 1, 3 and 12 (Table 4). EPT predicted SFCT at month 3 (Table 5).

Table 4 Multiple linear regression analysis of postoperative CMT change compared with baseline
Table 5 Multiple linear regression analysis of postoperative SFCT change compared with baseline

Multiple Linear Regression Analysis of the FLACS Group

In the FLACS group, baseline CMT values predicted CMT at week 1, month 1 and month 3 (Table 4). Ametropia predicted SFCT at month 12 (Table 5).


Ocular microcirculation dysfunction caused by diabetes mellitus can affect the retinal blood supply, causing hypoxia and destruction of the retinal barrier. Even in NoDR patients who have no significant decrease in vision and no observable changes in diabetic retinopathy at the fundus, auxiliary examination may still reveal that the blood flow of the macula and the optic disc has changed [24]. Inflammatory reactions caused by cataract surgery, ultrasound power and the intraoperative photostress from microscopic light will affect the fundus of these patients. Multiple studies have shown that CMT in diabetic patients increases after PE surgery [25,26,27]. In particular, the incidence of macular edema in patients with moderate and severe NPDR increased significantly [28].

However, existing studies have confirmed that the microcirculation disorder caused by diabetes mellitus is not limited to the retina, and also affects the choroidal microcirculation, causing the choroidal blood supply to decrease [29, 30], and the choroid thickness decreases compared with that in normal individuals [31,32,33]. The choroid has the role of providing nutrition and oxygenation to the retinal pigment epithelium layer and the outer retina layer. The dysfunctional choroid also affects the retinal function and plays an important role in the development of DR [17]. Therefore, the use the retina and choroid parameters to observe fundus changes in diabetic patients after cataract surgery is more comprehensive. Because of the dense choroidal vessels in the fovea, SFCT was used as the choroid evaluation parameter and CMT was used as the macular evaluation parameter in this study. In order to ensure safety, the study excluded patients with moderate to severe NPDR and PDR whose microvascular damage was more severe.

The FLACS surgical method has the advantages of assisting capsulotomy and fragmentation, and shortening the surgeon's learning curve. Nagy et al. confirmed that although edema of the outer nuclear layer of the macula occurred after surgery with both methods, it was lighter in the femtosecond laser group [34]. In addition, no change was observed in choroidal thickness at month 1 after FLACS compared with the baseline [35]. However, femtosecond laser pulse shockwave can cause mechanical damage to anterior segment structures such as the iris and ciliary body, and microbubbles generated by femtosecond laser action can also mediate the inflammatory cascade. Compared with conventional PE surgery, FLACS led to a more significant increase in prostaglandin levels [14]. In addition, the melanosomes of the retinal pigment epithelium layer have a strong absorption effect on near-infrared laser light, and this effect may cause damage to the retina and choroid. Whether the increase in intraocular pressure caused by FLACS affects the blood supply of the retina and optic nerve has been a focus of research.

At present, there are many studies focused on the changes in retinal and choroidal thickness in diabetic patients after PE operation. Many scholars are skeptical about whether the FLACS application in diabetic patients is safe and effective, and whether it affects the retina and choroid. Therefore, this study will expand this discussion.

BCVA was effectively improved in both groups from week 1 after surgery, but a statistically significant difference in BCVA between the two groups was observed only in the early postoperative period. The cumulative dissipated energy (CDE) in the FLACS group was significantly lower than that in the PE group (P < 0.0001), which means less damage to surrounding structures and shorter recovery time [36, 37]. Wei et al. [38] used the Corvis ST to confirm that the effect of FLACS on corneal biomechanics was smaller than that of PE at week 1 and month 1. A prospective randomized intraindividual cohort study also confirmed that FLACS can lower the risk of corneal edema [39]. This may be the reason that BCVA was better in the FLACS group than in the PE group in the early postoperative period. A case–control study by Cavallini et al. [40] confirmed that a significant improvement in average BCVA was observed at month 3, but there was no statistical difference between the two groups. A meta-analysis by Kolb et al. suggested that visual acuity did not show any difference between two groups at month 1 and month 3, and these results were not believed to carry any clinical importance [41]. Another randomized controlled trial by Day et al. [42] confirmed that the average BCVA difference between PE and FLACS was −0.01 logMAR (P = 0.34) at month 3 after surgery. The study by Ewe et al. confirmed that at 6 months after surgery, FLACS did not demonstrate a clinically meaningful improvement in visual outcomes over PE surgery [43]. Day et al. [44] confirmed that binocular corrected distance visual acuity (CDVA) favoring FLACS (P = 0.036), although statistically significant, was not clinically important.

None of the patients in the two groups of the current study experienced serious intraoperative complications such as vitreous hemorrhage or posterior capsule rupture. This may be because the surgeon in this study was experienced in cataract surgery and the LenSx Laser system. For inexperienced doctors, this may be different [45]. At the same time, the results of this study showed that the CMT and SFCT in the FLACS group recovered to the preoperative level 12 months after surgery, which also confirmed that for NoDR patients or patients with mild NPDR, FLACS is safe in its effects on retinal and choroidal thickness.

After conventional PE surgery, the risk of macular thickening in diabetic patients is increased [46, 47]. Existing studies have reported that the central macular thickness increases in patients with mild and moderate NPDR in months 1, 3 and 6 after conventional PE [48], and there is no significant difference in the central macular thickness between baseline and post-operation in NoDR patients [49]. In this study, the postoperative CMT of both groups began to increase from week 1 after the operation and began to decline at month 3. At month 6, neither group had returned to preoperative levels. This is consistent with the results of Liu et al. [48]. However, after follow-up to month 12, we found that the CMT in the FLACS group recovered to the preoperative level, while the CMT in the PE group had a tendency to rise again.

Postoperative inflammation caused by tissue damage from cataract surgery, breakdown of the blood–retinal and blood–aqueous barriers, and the release of prostaglandins and vascular endothelial growth factor (VEGF) are all possible mechanisms causing macular thickening [50, 51]. Diabetic retinopathy can lead to increased permeability of retinal blood vessels and swelling of the macula [52]. Results reported by Vujosevic et al. [53] and Pierro et al. [33] confirmed the occurrence of dilated retinal superficial capillaries and pathological changes before cataract surgery in diabetic patients. Therefore, in such patients, cataract surgery has a higher risk of macular thickening. In this study, the CMT of the FLACS group recovered to the preoperative state at month 12, but that in the PE group did not. This may prove that the surgical impact and the postoperative inflammation caused by FLACS can be resolved in a shorter time.

Regression analysis shows that the baseline CMT was the most important predictor affecting the PE group and the FLACS group within 12 months and 3 months after surgery, respectively. Therefore, for diabetic patients, the preoperative CMT level should be considered. Since the fundus capillaries of diabetic patients who have not undergone DR fundus changes can be affected, for patients with high preoperative CMT levels, whether undergoing traditional PE surgery or FLACS, close attention should be paid to the postoperative retinal thickness and retinal capillary morphology and perfusion situation. Clinicians should be alert to the rapid development of diabetic retinopathy, obvious retinal thickening and even the appearance of macular edema.

Ibrahim et al. [54], using EDI-OCT, found that SFCT was significantly increased 1 week after PE and returned to the preoperative level at month 3 after surgery. However, conflicting results have been reported in studies on postoperative SFCT changes in diabetic patients, with some studies showing that SFCT increased after PE surgery in diabetic patients [55, 56], while others have found no change [57, 58], and yet others have shown a postoperative reduction in choroidal thickness [17]. In the present study, the SFCT increased at week 1 after surgery in the PE group and began to decline at month 3, but did not return to preoperative levels until month 12. Although a similar increase was seen in the early stage in the FLACS group, the SFCT had returned to preoperative levels 6 months after surgery.

Choroidal thickening is also related to surgical inflammatory response, the release of prostaglandins and cytokines, and the destruction of the blood–retinal barrier [59, 60]. The earlier recovery of SFCT to preoperative levels also confirmed that FLACS has a smaller effect on intraocular inflammation. The difference in EPT between the two groups was significant, and regression analysis showed that EPT did affect the SFCT relative to baseline in month 3 in the PE group, indicating that effective control of EPT can lead to faster SFCT recovery to preoperative levels, and FLACS has an intuitive and effective role in reducing EPT.

In order to prevent myosis caused by anterior chamber inflammation as a result of femtosecond laser surgery and consequent implications for subsequent surgical operations, the FLACS group in this study was given nonsteroidal anti-inflammatory eye drops before surgery. Studies have confirmed that postoperative use of nonsteroidal anti-inflammatory drugs will slow the increase in retinal and choroidal thickness [61]. Therefore, whether preoperative prophylactic use has an effect on postoperative retinal and choroidal thickness requires further study.

Our research has certain limitations. Firstly, the sample size was small and surgery was not randomly assigned to patients. The choroidal thickness was measured manually at a single point. Furthermore, the study did not include patients with moderate to severe NPDR and PDR proliferative retinopathy. Related mechanisms still need further study.


In our sample population, compared with traditional PE surgery, FLACS effectively reduced EPT, achieving good vision earlier and earlier recovery of retinal and choroidal thickness. Baseline CMT was found to be an important predictor of postoperative CMT changes in FLACS and traditional PE surgery. Further studies are needed to confirm these findings.