Incidence of posterior vitreous detachment after three different types of laser refractive surgery

Abstract

Purpose

To compare the incidence of posterior vitreous detachment (PVD) after small incision lenticule extraction (SMILE), femtosecond laser in situ keratomileusis (FS-LASIK), and transepithelial photorefractive keratectomy(T-PRK), and to evaluate its potential risk factors and association with early postoperative visual and refractive outcomes.

Methods

Patients who met the eligibility criteria for SMILE, FS-LASIK, or T-PRK after comprehensive preoperative evaluation were enrolled. B-scan ultrasonography was performed by a single operator one day before surgery, and patients with preoperative PVD were excluded. Finally, 24 patients (45 eyes) were included in the SMILE group (Group S), 21 patients (39 eyes) in the FS-LASIK group (Group F), and 22 patients (44 eyes) in the T-PRK group (Group T). All procedures were performed by the same surgeon. B-scan ultrasonography was repeated by the same operator 1 week after surgery to detect newly developed PVD.

Results

The mean age was 21.5 ± 5.56 years, 23.9 ± 5.00 years, and 22.9 ± 4.56 years in Groups S, F, and T, respectively (p = 0.484). The mean axial length was 25.80 ± 0.87 mm, 25.91 ± 0.80 mm, and 25.72 ± 0.94 mm, respectively (p = 0.603), and the mean spherical equivalent was −4.65 ± 1.49 D, −5.33 ± 1.79 D, and −4.70 ± 1.36 D, respectively (p = 0.080). Postoperative PVD was detected in 9 eyes in Group S (20.0%; 95% CI, 9.6%–34.6%), 7 eyes in Group F (17.9%; 95% CI, 7.5%–33.5%), and 5 eyes in Group T (11.4%; 95% CI, 3.8%–24.6%) (p = 0.570).

Conclusions

No statistically significant difference in early postoperative PVD incidence was observed among SMILE, FS-LASIK, and T-PRK within one week after surgery. Further studies with larger sample sizes and longer follow-up are needed to confirm these findings.

Background

The number of people with myopia is increasing worldwide, and refractive surgery has become an important method for correcting myopia in recent years. Currently, the most widely used laser refractive surgery methods are small incision lenticule extraction (SMILE), femtosecond laser in situ keratomileusis (FS-LASIK), and transepithelial Photorefractive keratectomy (T-PRK) [1]. These three procedures have shown favorable safety and efficacy [2]. However, less common complications may still occur, including posterior vitreous detachment (PVD) [3], rhegmatogenous retinal detachment (RRD) [4], macular hemorrhage, and macular perforation [5].

PVD is a predisposing factor for retinal detachment and deserves attention from physicians performing refractive surgery [6]. Previous studies have evaluated PVD after microkeratome-assisted LASIK and FS-LASIK, with reported incidences ranging widely from 9% to 85% [7, 8]. In addition, one study found no significant difference in PVD incidence between LASIK and PRK groups [9]. Refractive surgery has developed rapidly over the last two decades. SMILE was introduced in 2011 [10], and has since been increasingly adopted worldwide. Compared with FS-LASIK, SMILE uses a femtosecond laser alone to complete refractive correction, potentially reducing surgical time and procedure-related side effects [11–13]. However, data on the occurrence of PVD after SMILE remain limited.

Therefore, this study aimed to compare the incidence of early postoperative PVD among patients undergoing SMILE, FS-LASIK, and T-PRK. In addition, we preliminarily explored whether early postoperative PVD detected at one week was associated with short-term visual acuity and refractive outcomes.

Methods

This was a prospective, non-randomized, comparative study. Patients were recruited from the Affiliated People’s Hospital of Ningbo University between February and August 2024. This study was approved by the Research Ethics Committee of the Affiliated People’s Hospital of Ningbo University. Data collection conformed to all local laws and complied with the principles of the Declaration of Helsinki.

All patients underwent a comprehensive preoperative evaluation by a refractive surgeon to determine their eligibility for refractive surgery. The evaluation included slit-lamp examination of the anterior segment, intraocular pressure measurement, corneal biomechanical assessment, corneal topography, fundus examination, and other necessary ophthalmic examinations. The surgeon then selected the most appropriate surgical procedure for each patient based on the preoperative findings. B-scan ultrasonography was performed by a single operator using an E-Z Scan AB5500+ device (Sonomed Inc, New Hyde Park, NY, USA). After applying the topical gel, the probe was pressed against the lightly closed upper eyelid using a high-to-maximum gain to detect PVD. A kinetic examination was performed, and PVD was diagnosed based on the presence of a low-reflectivity mobile echogenic membrane that was partially or completely detached from the retina. Patients who already had a PVD were excluded from the study. For eyes in which PVD status was difficult to determine, the images were reviewed by two additional experienced ultrasound physicians. The inclusion criteria were as follow: 1) myopia less than 8.00 D, 2) age between 18 to 35 years, 3)eligibility for refractive surgery based on the comprehensive preoperative evaluation described above, 4) Ocular Surface Disease Index (OSDI) scores ≤ 13 points, 5) no history of previous ocular surgery or ocular trauma, 6) no preoperative PVD. Patients were excluded if they had ocular pathological conditions, relevant systemic diseases, OSDI score > 13 points, a history of ocular surgery or trauma, or preoperative PVD. The included patients were classified into three groups according to the surgical procedure: Group S for SMILE, Group F for FS-LASIK, and Group T for T-PRK.

All patients underwent refractive surgery on the following day. All the surgeries were performed by an experienced surgeon. A VisuMax femtosecond laser (Carl Zeiss Meditec AG, Jena, Germany) was used for SMILE. It involved removing a stromal lenticule, with a 110 µm-thick cap, the negative pressure is between −65kpa to −78kpa, and the suction time is 25s from the beginning of suction application until suction release. In FS-LASIK, a lamellar flap with a superior hinge was created using the same femtosecond laser (Carl Zeiss Meditec AG, Jena, Germany). The maximum thickness ranged from 90 to 110 µm. Tissue ablation was performed using the Amaris laser. the negative pressure is between −65kpa to −78kpa, and the suction time is 15s from the beginning of suction application until suction release. T-PRK was performed using the aberration-free mode of an Amaris 750 s excimer laser (Schwind eye-tech-solutions, Kleinostheim, Germany) with smart-pulse technology, and the cornea was ablated in a single step. After SMILE and FS-LASIK, 0.1% fluorometolone and artificial tear drops were used at 4 times a day to surgical eyes in the first week and then 0.1% fluorometolone was gradually reduced once a week. One month after surgery, the patient did not use any eye drops, except for artificial tear drops. After T-PRK, Tobramycin Dexamethasone and artificial tear eye drops were used in the first week for 4 times a day, and then switched to 0.1% fluorometholone eye drops for 4 times, then 0.1% fluorometolone was gradually reduced once a month. Four months after the surgery, no eye drops other than artificial tears were used.

At one week postoperatively, intraocular pressure measurement, visual acuity testing, refraction, and B-scan ultrasonography were performed for all patients. B-scan ultrasonography was performed by the same operator using the same technique as that used preoperatively to detect PVD. The operator was blinded to the type of surgery; for some eyes in which it was difficult to determine whether PVD had occurred, the images were reviewed by two additional experienced ultrasound physicians. Eyes were included in the PVD group only when all three physicians agreed that PVD had occurred. The images were evaluated in five orientations: central, superior, inferior, nasal, and temporal. If new PVD was detected, its location was recorded. Visual acuity, intraocular pressure, and refraction were also recorded.

Statistical analyses were performed with SPSS 24 (Statistical Package for the Social Sciences version 24 (SPSS Inc., Chicago, IL, USA). Continuous variables were expressed as mean ± standard deviation (SD), whereas categorical variables were expressed as n (%). Comparisons between the different parameters in the three groups were performed using ANOVA. The incidence of PVD in the three groups was compared using the Fisher’s exact test. p < 0.05 was considered statistically significant.

Results

The SMILE, FS-LASIK, and T-PRK groups comprised 24 (45 eyes), 21 (39 eyes), and 22 (44 eyes) patients, respectively. Baseline patient characteristics are shown in Table 1. In total, 15 patients (21 eyes) developed new PVD after surgery. Specifically, new PVD developed in 7 patients (9 eyes) in Group S, 5 patients (7 eyes) in Group F, and 3 patients (5 eyes) in Group T. Among these 15 patients, 8 were male and 7 were female. All 7 patients in Group S were male, 4 of 5 patients in Group F were female, and all 3 patients in Group T were female. Among the 21 affected eyes, 8 were right eyes and 13 were left eyes. Specifically, 3 eyes in Group S, 3 eyes in Group F, and 2 eyes in Group T were right eyes.

Table 1.

Demographic data of three groups

	Group S	Group F	Group T	P
Male/Female	20/4	11/10	14/8	0.08
Mean age±SD (y)	21.5 ± 5.56	23.9 ± 5	22.9 ± 4.56	0.484
Mean axial length±SD (mm)	25.8 ± 0.87	25.91 ± 0.8	25.72 ± 0.94	0.603
Mean Spherical equivalent±SD (D)	−4.65±1.49	−5.33±1.79	−4.70±1.36	0.08

p < 0.05 was considered statistically significant

PVD location was classified as nasal, temporal, superior, inferior, or central. The location of PVD was carefully recorded in all affected eyes. PVD was detected in the nasal region in 12 eyes, the temporal region in 11 eyes, the central region in 4 eyes, the superior region in 2 eyes, and the inferior region in 2 eyes. Only one eye developed complete vitreous detachment (Figs. 1 and 2).

Fig. 1.

B-scan ultrasonography was performed before surgery (no PVD)

Fig. 2.

B-scan ultrasonography was performed 1 week after surgery (red arrow represents PVD)

The overall incidence of PVD in this study was 16.4%(95% CI, 10.5%–24.0%). The incidence of postoperative PVD was 20.0% (9/45; 95% CI, 9.6%–34.6%) in the SMILE group, 17.9% (7/39; 95% CI, 7.5%–33.5%) in the FS-LASIK group, and 11.4% (5/44; 95% CI, 3.8%–24.6%) in the T-PRK group. No statistically significant difference was observed among the three groups (p = 0.570).

During the postoperative three months period, all patients recovered well postoperatively (Fig. 3), and no serious posterior segment complications occurred.

Fig. 3.

Visual acuity and diopter changes during the whole follow-up

Discussion

After SMILE and FS-LASIK, new PVD developed in 7 patients (9 eyes) and 5 patients (7 eyes), respectively. This may be related to suction application during SMILE and FS-LASIK. Previous studies have suggested that the onset and release of suction can cause rapid increases and decreases in intraocular pressure (IOP), resulting in changes in anterior segment morphology, displacement of the lens and vitreous, and eventually PVD [14]. In addition, application of the suction ring may lead to axial elongation, resulting in vitreoretinal traction at vulnerable sites such as the macula and vitreous base, which may contribute to posterior segment changes, including PVD [15].

However, new PVD also occurred in 5 eyes after T-PRK, although suction was not used during this procedure. This may be related to the propagation of excimer laser-induced shock waves. Pulsed laser ablation of any type of material can generate a stress wave at the surface of the object that propagates into the tissue [16]. Therefore, excimer laser-induced shock waves may represent a potential iatrogenic factor that generates stress in the posterior segment of the eye.

Although the difference was not statistically significant, PVD was numerically more frequent in the SMILE group than in the FS-LASIK group. Both SMILE and FS-LASIK involve suction application during surgery, and the negative pressure levels were identical in the present study. This numerical difference may be related to the longer suction time during SMILE. The suction time was 25 s during SMILE and 15 s during FS-LASIK. The cumulative effect of suction duration may have contributed to the relatively higher incidence of PVD after SMILE, which is consistent with previous findings [17].

The incidence of postoperative PVD after FS-LASIK in our study was 17.9%, which is consistent with the study by Gavrilov et al. (2011) [7], but differs substantially from the study by Osman et al. [17], who reported a postoperative PVD incidence of 85% (17/20) after FS-LASIK. This discrepancy may be partly explained by differences in the timing of postoperative B-scan assessment: PVD was assessed 1 week after FS-LASIK in our study, whereas it was assessed 1 month postoperatively in their study. A longer follow-up interval may capture PVD events that are less directly related to the surgical procedure. In addition, Osman et al. used a different femtosecond laser platform to create the flap (IntraLase FS-150; Abbott Medical Optics, Santa Ana, CA, USA). The mean suction time in their FS-LASIK group was 63 ± 4 s, which was much longer than that in our FS-LASIK group (approximately 15s). The cumulative effect of suction duration may therefore partly explain the difference in PVD incidence. Furthermore, in our study, eyes with ambiguous findings were classified as having PVD only when all three physicians agreed that PVD had occurred, which may also have contributed to differences from previous reports.

In addition to evaluating the incidence of early postoperative PVD, we further assessed whether early PVD was associated with postoperative visual and refractive outcomes. During the 3-month follow-up period, favorable visual and refractive outcomes were observed in all patients, and no serious retinal complications occurred. This finding may be clinically reasonable because uncomplicated early PVD mainly involves changes at the vitreoretinal interface, whereas visual and refractive outcomes after corneal laser refractive surgery are primarily determined by corneal refractive correction, corneal wound healing, corneal biomechanical changes, and postoperative refractive stability [18, 19]. Therefore, in the absence of retinal or macular complications, early postoperative PVD alone may have limited influence on routine visual acuity and refraction outcomes during the early postoperative period.

Nevertheless, PVD remains clinically relevant. Previous studies have reported that PVD may be associated with symptoms such as floaters and photopsia, and in some cases may lead to traction-related complications, including retinal tears, retinal detachment, vitreous hemorrhage, macular hole, or epiretinal membrane formation [20–22]. These complications, rather than uncomplicated PVD itself, are more likely to affect visual function. Therefore, careful fundus evaluation and appropriate postoperative follow-up remain important, particularly in patients with new-onset floaters, photopsia, or other vitreoretinal symptoms.

This study has some limitations. First, the number of patients in each group was relatively small. A larger sample size is needed to determine whether differences in PVD incidence exist among the three procedures.

The short follow-up period represents another limitation of this study. PVD is a dynamic and progressive process that may evolve over weeks, months, or even years. In the present study, postoperative PVD assessment was performed only at one week after surgery. This time point was selected to capture early postoperative changes that might be more directly and temporally related to the surgical procedure, while reducing the potential influence of later confounding factors, including age-related vitreous degeneration, natural progression of myopia-associated vitreoretinal changes, ocular trauma, postoperative behavioral factors, or other unrelated ocular events. Previous studies have assessed early PVD changes after refractive surgery at similarly short postoperative intervals, including 48 hours, one week, and one month after surgery [7, 8, 17, 20]. Future studies with longer follow-up intervals are warranted to more comprehensively characterize the temporal pattern of PVD development after surgery.

Another limitation of this study is that PVD was diagnosed using B-scan ultrasonography rather than OCT [23, 24]. Although B-scan ultrasonography is widely used in clinical practice and allows dynamic evaluation of the posterior vitreous, its ability to detect subtle vitreoretinal separation may be limited by its relatively lower resolution and operator dependence. In early, focal, or partial PVD may be difficult to identified by B-scan ultrasonography alone. OCT has substantially improved visualization of the vitreoretinal interface and is increasingly considered a preferred imaging modality for detecting and staging early or partial PVD [25–28]. Uchino et al. demonstrated that OCT could identify the initial stages of PVD, including incomplete perifoveal PVD with persistent attachment to the fovea and optic disc [25]. Subsequent studies comparing ultrasonography and OCT in characterizing PVD status, and have further highlighted the value of OCT [26, 27]. More recently, widefield swept-source OCT(SS-OCT) has enabled broader visualization of the posterior vitreous cortex and quantitative evaluation of PVD stages [28]. Therefore, the absence of standardized OCT data in the present study may have led to underestimation of early or partial PVD. Future prospective studies should incorporate standardized OCT-based PVD staging, preferably using widefield or SS-OCT in combination with B-scan ultrasonography, to more accurately evaluate the onset, progression, and clinical relevance of PVD.

Conclusions

In conclusion, this study found no significant difference in early postoperative PVD incidence among SMILE, FS-LASIK, and T-PRK within one week after surgery. Further prospective studies with larger cohorts and longer follow-up are required to evaluate the medium- and long-term vitreoretinal safety of these three surgery procedures.

Abbreviations

PVD

posterior vitreous detachment

SMILE

small incision lenticule extraction

FS-LASIK

femtosecond laser in situ keratomileusis

T-PRK

transepithelial photorefractive keratectomy

RRD

rhegmatogenous retinal detachment

Author contributions

Study conception and design (WWH, ML and LH); Data acquisition and analysis (WWH, XMZ, LY); Manuscript writing (WWH, ML); Manuscript polishing (LM, LQK) Critically revising the manuscript for intellectual content: (LH). All authors made final approval of the version to be published and agreed to be accountable for all aspects of the work.

Funding

1. Ningbo Clinical Research Center for Ophthalmology (2022L003). 2. Ningbo Key Laboratory for neuroretinopathy medical research. 3. the Project of NINGBO Leading Medical & Health Disipline (2016-S05). 4. Ningbo “Innovation Yongjiang 2035” Key Technology Breakthrough program (2024Z233).

Data availability

Data is provided within the manuscript, the datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Prior to the study, all participants were fully informed of the research content and potential risks involved, and written informed consent was obtained. The study protocol and the informed consent form were approved by the Ethics Committee of The Affiliated People’s Hospital of Ningbo University (approval no. 2024–064) and complied with the tenets of the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.