Refractive Outcomes following Vitrectomy and Phacovitrectomy in Rhegmatogenous Retinal Detachment


 Background

Cataracts are commonly seen during the diagnosis of rhegmatogenous retinal detachment (RRD) or occur following its surgical management. Refractive changes following combined phacoemulsification and vitrectomy (phacovitrectomy) can affect the quality of vision. This study aimed to evaluate the refractive changes following vitrectomy versus phacovitrectomy in patients with RRD.
Methods

A total of 96 patients who underwent retinal surgery for RRD in one eye, between August 2016 and January 2018, were enrolled and divided into three groups. In 18 eyes, 23-gauge transconjunctival sutureless vitrectomy (23G TSV) was performed (group A); 25G TSV was performed in 49 eyes (group B), and phacovitrectomy with 25G TSV was performed in 29 eyes (group C). Silicone oil was injected after the surgery in all three groups. Anterior and posterior corneal astigmatism, axial length (AL), and anterior chamber depth (ACD) were measured at baseline and postoperatively at 1 week, 1 month, and 3 months.
Results

Anterior and posterior corneal astigmatism did not change, and AL continued to increase for 3 months after surgery, whereas ACD decreased postoperatively but quickly returned to baseline in groups A and B. In group C, anterior and posterior corneal astigmatism, AL, and ACD increased postoperatively, and corneal astigmatism returned to baseline by 3 months, but AL and ACD continued to increase for 3 months.
Conclusion

Vitrectomy alone does not affect anterior or posterior corneal astigmatism. Following vitrectomy, the AL increases, while the ACD decreases transiently. When vitrectomy is combined with phacoemulsification for cataract removal, the anterior and posterior corneal astigmatism increases transiently, while AL and ACD are enlarged for several months following surgery.


Results
Anterior and posterior corneal astigmatism did not change, and AL continued to increase for 3 months after surgery, whereas ACD decreased postoperatively but quickly returned to baseline in groups A and B.
In group C, anterior and posterior corneal astigmatism, AL, and ACD increased postoperatively, and corneal astigmatism returned to baseline by 3 months, but AL and ACD continued to increase for 3 months.

Conclusion
Vitrectomy alone does not affect anterior or posterior corneal astigmatism. Following vitrectomy, the AL increases, while the ACD decreases transiently. When vitrectomy is combined with phacoemulsi cation for cataract removal, the anterior and posterior corneal astigmatism increases transiently, while AL and ACD are enlarged for several months following surgery.

Background
Loss of visual function in rhegmatogenous retinal detachment (RRD) can be prevented by prompt vitreoretinal surgical intervention. In geriatric patients presenting with vitreoretinal disorders, cataracts are often a comorbid diagnosis. In the absence of cataracts, the vitreoretinal surgery, together with silicon oil lling, could accelerate the process of cataractogenesis to some extent. Thus, combined phacoemulsi cation and vitrectomy has been advocated as an effective and safe procedure in the management of patients with both cataract and posterior segment disease [1]. However, refractive changes following such surgeries can affect the quality of vision and cannot be ignored. Therefore, our study aimed to evaluate the refractive outcomes following vitrectomy versus phacovitrectomy in patients with RRD.

Patients
This retrospective study analyzed the data of a total of 96 patients with RRD who underwent vitrectomy or phacovitrectomy, between August 2016 and January 2018. In 18 eyes, 23-gauge transconjunctival sutureless vitrectomy (23G TSV) was performed, while 25G TSV was performed in 49 eyes, and phacoemulsi cation combined with 25G TSV (phacovitrectomy) was performed in 29 eyes. Silicone oil was injected in all eyes at the end of surgery.

Procedures
Surgery was performed under local or general anesthesia by a single surgeon. The standard three-port technique without scleral buckling was used for vitrectomy in all eyes, of which 18 eyes underwent 23G TSV, and 49 eyes underwent 25G TSV. The retinal holes were closed by laser. After the fluid-air exchange, silicone oil was injected to replace the air entirely and fill the vitreous cavity.
For 29 eyes that underwent phacovitrectomy, the cataract removal was performed rst. A 3.2 mm scleral incision was made superiorly with two side-ports at the 10 and 2 o'clock positions. A continuous curvilinear capsulorhexis was created, followed by phacoemulsi cation and aspiration of the crystalline lens. Bleeding points were electrocoagulated. The three-port 25G TSV was then performed. All retinal holes were closed using the laser. Finally, the intraocular lens (IOL) was implanted in the capsular bag.
The scleral incision was sutured using 10-0 polyamide under minimal tension, and silicone oil was injected. AL was measured by AL Scan (NIDEK, Japan). Anterior and posterior corneal astigmatism and ACD were measured by Pentacam (OCULUS, Germany).

Statistical analysis
Statistical Package for the Social Sciences (SPSS) software Version 21.0 SPSS Inc., Chicago, IL, USA) was used for all statistical analyses. All data are represented as the mean value ± standard deviation (mean ± SD). Differences between groups were analyzed by repeated measures. When statistically signi cant, the Least Signi cant Difference (LSD) was further calculated. P < 0.05 were considered statistically signi cant.

Results
A total of 96 patients were enrolled and divided into three groups. In 18 eyes, 23G TSV was performed (Group A); 25G TSV was performed in 49 eyes (Group B), and phacovitrectomy in 29 eyes (group C).

Anterior corneal astigmatism
Changes in anterior corneal astigmatism following surgery occurred only in group C ( Table 1). The anterior corneal astigmatism increased signi cantly from baseline (1.73 ± 0.56, P < 0.05) to 2.28 ± 0.64 at 1 week, and 2.23 ± 0.77 at 1 month. At postoperative month 3, the anterior corneal astigmatism receded back (1.75 ± 0.65) to baseline. Fig. 1 shows the trends in corneal anterior astigmatism changes at different time points for all three groups. Posterior corneal astigmatism Following surgery, the posterior corneal astigmatism in group C increased signi cantly at 1 week (0.63 ± 0.18) and 1 month (0.59 ± 0.15) follow up, compared to baseline (0.40 ± 0.13, P < 0.05). No such changes were observed in groups A and B (Table 2). Fig. 2 shows the trends of posterior corneal astigmatism in three groups at different time points before and after surgery. Three months following surgery, the increased posterior corneal astigmatism in group C decreased to baseline. No signi cant difference in corneal astigmatism was observed between the three groups at baseline or 3 months after surgery (P > 0.05). Data are presented as mean ± SD unless otherwise indicated *P < 0.0 compared with baseline; † P < 0.05 compared with Group A; ‡ P < 0.05 compared with Group B

Axial length
Compared with baseline, there was a signi cant increase in AL at 1 week, 1 month, and 3 months (P < 0.05) following surgery in all three groups (Table 3). Fig. 3 shows the trends of axial length at different times for the three groups. No signi cant differences were observed between the three groups at the baseline, 1 month, and 3 months after surgery. At 1 week, the AL in group C (26.05 ± 0.61, P < 0.05) increased signi cantly compared to group A (25.69 ± 0.50) and group B (25.64 ± 0.56). compared to baseline (2.75 ± 0.14, P < 0.05). In group C, the ACD increased signi cantly following surgery and remained at the same level throughout the follow up (Table 4). Fig. 4 shows the ACD at different times for the three groups. Data are presented as mean ± SD unless otherwise indicated *P < 0.05 compared with baseline; † P < 0.05 compared with Group A; ‡ P < 0.05 compared with Group B Statistically signi cant differences were found among all three groups at week 1 (P < 0.05), with the most pronounced ACD increase in group C (3.02 ± 0.16) followed by group A (2.58 ± 0.17). The least increase was seen in group B (2.46 ± 0.15). During follow up at months 1 and 3, the ACD increase was more obvious in group C than in groups A or B.

Discussion
Currently, vitrectomy is commonly combined with cataract surgery in clinical practice for the treatment of both RRD and cataracts. The refractive changes following surgery play an important role in the quality of vision. The size of vitrectomy incision, suturing of vitrectomy ports, intraocular llers (type, nature, density, and volume), and the size and suturing of cataract incision in uences the postoperative refractive status of the eyeball to varying degrees.
Pentacam is an accurate modality for measuring corneal changes [2]. In the past, the A-scan ultrasound was used to measure the distance between the anterior surface of the cornea and the inner limiting membrane of the retina to measure AL. The AL Scan is an optical biometer that measures the distance between the anterior surface of the cornea and the retinal pigment epithelium with a superior accuracy compared to the A-scan ultrasound. Since the AL Scan is based on the principle of partial coherence interferometry, the results were not affected by silicone oil, and there was good agreement between the AL Scan and IOLMaster in measuring the AL [3][4][5]. Thus, we justify the use of Pentacam for measuring corneal astigmatism and ACD and AL Scan for measuring AL.
Mechanisms in post-surgical corneal astigmatism Postoperative corneal astigmatism that appears following vitrectomy arises mainly from the scleral threeport incision and the physical traction of suturing. These factors inconsistently change the corneal refractive power in every diameter, resulting in the formation of corneal astigmatism [6][7][8]. With the development of minimally invasive vitrectomy techniques in recent years, the diameter of the vitreouscutting head has become thinner, making the surgery "sutureless." Some groups reported that sutureless 23G and 25G vitrectomy eliminated the inconsistent changes of corneal refractive power in every diameter [9]. Park DH et al. found that during 23G TSV with silicone oil lling, the cornea could still undergo a transient change in refraction resulting mainly from the tension and lling volume of silicone oil [10].
In combined surgery (phacovitrectomy), the main reason for corneal astigmatism was cataract surgery, especially the thermal injury caused by phacoemulsi cation and the sutures after operation [11]. Electrocoagulation of bleeding vessels could cause shrinkage of scleral collagen bers, resulting in wound contraction. This not only changed the corneal curvature but also led to a steeper cornea at the corresponding site and thus corneal astigmatism [12]. Sayed KM et al. described that a transscleral incision for phacoemulsi cation was associated with no signi cant effect on corneal astigmatism, especially in the absence of suturing, and only the corneal surface irregularity index increased in the early postoperative period, which usually returned to baseline after 6 months [13].
A larger and longer cataract surgical incision destroyed the structure of the corneal dome more severely, leading to a more signi cant postoperative corneal astigmatism [14]. Translimbal incision more likely causes corneal astigmatism than transscleral tunnel incision [15], and tighter sutures caused noticeable changes in corneal astigmatism [16].
In the current study, we eliminated the inconsistent changes of corneal refractive power in every diameter during 23G TSV and 25G TSV by making the procedure sutureless. This explains why both the anterior and posterior corneal astigmatism did not change signi cantly following surgery. In the phacovitrectomy group, the anterior and posterior astigmatism increased most obviously at post-surgical week 1 and returned to baseline by 3 months. The possible reasons being the thermal injury of electrocoagulation, and physical traction exerted by the sutures. Our ndings were similar to previous studies that reported cataract surgery as the main cause of corneal surface astigmatism in combined phacovitrectomy [11,12].

Mechanisms in uencing the post-surgical axial length
During phacovitrectomy, the change in AL is mainly caused by vitrectomy since cataract surgery had no signi cant effect on AL [17]. The suturing of the three-port incision shortens the scleral length in the axial direction, which in turn results in a shortening of the AL, with no apparent relationship to the properties of the llers [18]. Federman JL et al. believed that the reason for the postoperative increase in AL might be measurement errors. They used an A-scan ultrasound that measured the distance between the anterior surface of the cornea and the inner limiting membrane of the retina. Thus, the measurement was smaller than the real length before the recovery of the retina. But when the retina was reset, the AL measurement tended to be accurate, but longer [19]. Huang C et al. used the IOL master and similarly found that the AL increased after combined phacovitrectomy. However, they speculated that the physical property of llers was different from that of the vitreous and could not provide su cient support for the eyeball. With simultaneous lens removal, the eyeball deformed and tended to elongate [20][21][22]. Furthermore, the gasliquid exchange and the transient high intraocular pressure (IOP) during vitrectomy could also contribute to axial elongation [23]. Silicone oil, a commonly used ller, is a viscous substance, which could redistribute the liquid density of the vitreous cavity, producing a certain tension that increases the eyeball's AL [24]. Certain groups postulated that the combination of low preoperative IOP along with postoperative normal or high IOP could result in a long eyeball [17].
The AL changes were positively correlated with the viscosity and lling amount of silicone oil [23,25]. Also, the IOL does not provide enough brace to the eye compared to the original lens, so it was necessary to increase the amount of silicone oil to support the eyewall to some extent. This effect was transient and usually returns to baseline between a week to a month following surgery [20][21][22], after which period, the AL tended to be stable [26].
In our study, the AL of three groups increased the most in week 1, and these changes lasted until 3 months after surgery. We believe the main reasons for this AL increase were due to the tension exerted by silicone oil on the eyewall, as well as the transient high IOP of gas-liquid exchange. We eliminated the in uence of suturing and measurement error by making the surgery sutureless and by using the AL Scan for AL measurements. There was a minimal increase in AL at 3 months compared to 1 month, possibly due to head position, in ammation, and IOP uctuations. The phacovitrectomy group showed the most AL change at week 1, probably due to the temporary effect of silicon oil tension and lens removal.
Mechanisms in uencing post-surgical anterior chamber depth During vitrectomy, the ACD decreases due to the tension and viscosity of silicone oil and prone positioning, which displaces the iris-lens septum forwards [20,27]. Surgical trauma and scleral suturing could lead to ciliary body edema and effusion, resulting in a shallow anterior chamber [20]. ACD is also positively correlated with IOP, which means that a low postoperative IOP might result in a shallow anterior chamber [28].
In contrast, the ACD tends to increase during phacovitrectomy. The crystalline lens is disc-shaped with an anteroposterior diameter of about 5 mm, whereas the thickness of the IOL is approximately 1 mm, resulting in insu cient anterior segment support following cataract surgery. Also, the anterior segment pressure is eliminated after lens removal, which increases ACD. The effect of vitrectomy-related factors is relatively small compared to cataract surgery-related factors, and so the ACD tends to increase in combined phacovitrectomy [29].
When using silicone oil, a more substantial lling volume has a greater in uence on ACD [30]. Scleral suturing is more likely to cause a shallower anterior chamber than a sutureless incision [20]. A large scleral incision that is sutured tight causes a more pronounced ciliary body edema and has a more considerable in uence on ACD [20,31]. ACD showed the most obvious change during the rst postsurgical week. With wound healing and the stabilization of the lling and IOL, the ACD gradually returned to baseline by 2 to 4 months after surgery [1,32].
In our study, the ACD decreased the most during the rst week following 23G TSV and 25G TSV. The tension and viscosity of silicone oil and the prone patient positioning that promotes anterior displacement of the iris-lens septum could have caused the observed effect. Some degree of ciliary body edema and effusion might have in uenced the nal ACD. We found that with phacovitrectomy, the ACD increased due to IOL implantation and vitrectomy, as discussed previously. This change was most obvious at post-surgical week 1 and lasted until 3 months. ACD changes were more obvious in the phacovitrectomy group than the 23G TSV and 25G TSV groups. Besides, statistically signi cant differences between 23G TSV and 25G TSV groups were more pronounced in the 25G TSV group. This might be due to the small sample size of the 23G TSV group.
This study has a few limitations. First, this was a retrospective study with a small sample size per group. Second, results were obtained for only 3 months of follow up after the surgery. Third, since all measurements were performed before the removal of oil and sutures, the in uence of these factors on the refractive changes cannot be ruled out. Nevertheless, this study compared the refractive changes between some of the widely practiced new surgical techniques. We recommend large-sample, prospective studies with longer follow-up to accurately determine the changes in refractive error following vitrectomy versus phacovitrectomy.

Conclusions
Vitrectomy alone affects AL and ACD. Corneal astigmatism changes only if sutures are applied during vitrectomy. However, obvious changes occur in corneal astigmatism, AL, and ACD following combined phacovitrectomy. Speci c surgical steps minimize post-phacovitrectomy-refractive changes. For example, during phacovitrectomy, it is imperative to maintain a minimally invasive and gentle surgical approach to minimize post-surgical in ammation and scarring. The choice of an appropriate intraocular ller, its lling volume, and the degree of suture tightness in uence the refractive outcome. We recommend a moderate volume of ller and moderate suture tightness for optimal refractive outcomes. The study design was approved by the Institutional Human Experimentation Committee and Institutional Review Board (IRB) at the Second Hospital of Jilin University (3D5172173429) and adhered to the tenets of the Declaration of Helsinki. All subjects received details of the nature and possible consequences of the study, following which they provided their oral informed consent.They only needed do some examinations and there was no any harm,so they just provided their oral informed consent.

Consent for publication
All of the authors agreed to publish the article. All identifying information has been appropriately anonymized, and there is no identifying images used within the manuscript.
Availability of data and materials The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.