This was a retrospective study that included consecutive cases undergoing surgery for primary rhegmatogenous RD in patients with pseudophakia after phacoemulsification between the dates of January 2000 and June 2008 that had a follow-up duration of at least 12 months. Eyes were excluded from the study if they had non-primary rhegmatogenous RDs, such as from trauma, RDs associated with coloboma, uveitis, or glaucoma, RDs with a macular hole, or had an axial length >26 mm. Patients were excluded from the study if they had atopic dermatitis or if they had a pre-operative proliferative vitreoretinopathy grade C-1 or higher according to the Retina Society Classification [6].

The baseline data (gender, age, laterality, intraocular lens [IOL] status, best-corrected visual acuity, and intraocular pressure [IOP]) and the intra-operative findings (operative procedure, macular status, lens capsule status, extension of the RD, number of retinal breaks, and location of retinal breaks) were recorded. The visual acuity was recorded in Snellen units and converted to logarithm of the minimum angle of resolution (logMAR) units for statistical analysis. Each patient underwent a slit lamp biomicroscopy, fundus photography, and Goldmann's three mirror lens in order to identify retinal breaks at baseline. The duration of time from when the cataract surgery took place to the time at which the RD was diagnosed was recorded as the time of RD evolution. Scleral buckling (SB) or 20-gauge pars plana vitrectomy was performed according to the surgeon's preference. Vitrectomy combined with SB was considered for total RD, and multiple or unidentified retinal breaks.

Statistical analyses were performed using statistical software (SPSS ver. 11.0; SPSS Inc., Chicago, IL, USA). Univariate and multivariate linear regression analyses were undertaken for the dependent variable in each model. Univariate categorical analysis was performed using the two paired t-test, chi-square test, Mann-Whitney U-test, or the Fisher exact test, as appropriate. Data were analyzed via a repeated-measures analysis of variance with a Bonferroni correction. Correlation analyses were conducted via Spearman's correlation. Statistical significance was set at 0.05 (two-sided) for all of the tests.

This study comprised 106 eyes of 104 patients (70 males and 34 females), with a mean age of 55.4 ± 16.7 years (range, 20 to 85 years). The time between onset of RD (defined by the onset of early signs and symptoms, including flashes and floaters) and surgery was 15.4 ± 34.7 days (range, 4 to 75 days). The time of RD evolution was 35.8 ± 26.9 months (range, 1 to 120 months). The mean duration of follow-up was 18.0 ± 8.7 months (range, 12 to 60 months). Ten of the eyes (9.4%) have had previous ocular surgery (an anterior vitrectomy in six eyes, a penetrating keratoplasty in one eye and refractive surgery in two of the eyes).

Cataract surgery was performed on 16 eyes in our hospital and surgery was performed on the remaining 90 eyes in another hospital. There were no sclera tunnel incisions and no iris-fixated or anterior chamber lenses. A superior incision was made in four of the eyes (3.8%) and 60 eyes (56.5%) had a temporal incision. The incisional approach for 42 of these eyes (39.6%) was ill-defined. The location of the IOL was in the ciliary sulcus in six eyes, scleral fixation in four eyes, and in the bag in the remaining eyes.

Univariate analysis did not show any statistically significant differences between the baseline visual acuity and the following variables: gender; age; laterality; the time of RD evolution; the duration of symptoms; lower IOP; IOL location; lens capsule status; identification of a retinal break; number of retinal breaks; location of retinal breaks; and type of surgery. A smaller extension of the RD and a macula-on RD had better baseline visual acuity (p = 0.005, p = 0.002).

Surgical techniques in patients with pseudophakia after phacoemulsification have rapidly advanced in parallel with the development of better equipment. Previous studies involving patients with pseudophakic RDs have been reported before the era of phacoemulsification with micro-corneal incisions [5]. Thus, the reported surgical outcomes of patients with pseudophakic RDs may be improved upon. This retrospective study involved 106 patients with a primary of idiopathic pseudophakic rhegmatogenous RDs who had surgery by one surgeon over an 8-year period. Data relating to 21 pre-, intra-, and post-operative variables were studied.

The final reattachment rate was 82%, which is slightly greater than what previous studies have reported [6-8]. Previous reports have identified the following negative prognostic factors for reattachment after surgery: age >65 years; poor pre-operative vision; RD involving >3 quadrants; macula-off RDs; a lower IOP; inability to identify a retinal break; longer duration of symptoms; and grades C or D of proliferative vitreoretinopathy [6-13]. In the current study, only an inability to identify a retinal break was significantly associated with lower reattachment success. Proliferative vitreoretinopathy grade C or D was not analyzed due to the exclusion criteria used in this study.

A critical step in successful RD surgery is to locate and seal all of the causative retinal breaks. However, in pseudophakia, tiny breaks, poor mydriasis, cortical remnants, capsular opacification, glare or pitting from the intraocular lens implant, and corneal or vitreous opacities may make identification difficult, especially since anterior breaks commonly occur in pseudophakic and aphakic RDs. Thus, a retinal break is not identified in 9% to 20% of pseudophakic RDs, which would be a risk factor for surgical failure [11-17]. Cases with unidentified retinal breaks have traditionally been managed using SB and broad application cryopexy, which has a reported primary success rate of 53% to 70% [18,19]. We think an internal approach with vitrectomy offers some advantages that might result in the detection of these breaks. Scleral external indentation together with endoillumination and magnification during the vitrectomy may help locate previously undetected breaks [20]. Also during vitrectomy, breaks can be searched for with an aspirating instrument or by using perfluorocarbon liquids that may identify a stream of fluid that is exiting the subretinal space via the hole [21]. We performed vitrectomy combined with either segmental or encircled SB in all 30 of the eyes that had unidentified breaks with a final anatomical success rate of 66%. Previous case series have reported that the anatomical success rate of vitrectomy when combined with SB was superior to SB alone [19,22-24]. In contrast, some authors have found that vitrectomy was not of additional value in terms of a better success rate [25,26]. A larger prospective randomized trial is needed in order to determine an accurate surgical method for RDs with unidentified breaks.

As mentioned above, we indicated that identifying a retinal break is an important factor for both anatomic and functional outcomes for pseudophakic RD repair. The area of the retinal break was most frequently found in the superotemporal area. One of the most important changes in the posterior segment after cataract surgery is anomalous posterior vitreous detachment, which would induce a retinal tear [27-29]. Although the reasons for an increased incidence of posterior vitreous detachment after cataract surgery remain unknown, it is thought that the anterior movements of the vitreous during surgery might play a role in the development of posterior vitreous detachment [30,31]. We believe that it is possible that the insertion of an instrument during surgery via the main incision could produce vitreous movement and the subsequent vitreous traction may then result. In the current cases, the main incision for the cataract surgery was via a temporal approach and the eyes with an undetermined incisional approach may have undergone a temporal corneal approach. Although it is not clear why retinal breaks occur more frequently in the superiotemporal quadrant, meticulous examinations of the superiotemporal area during surgery are important to identify these retinal breaks.

It has been reported that the previously mentioned negative prognostic variables for anatomical outcomes are also associated with worse visual outcomes [5-13]. Among these variables, a lower IOP, unidentified retinal breaks, and the recurrence after primary surgery were associated with poorer final visual acuity in the current study. A lower IOP was related to a poor final visual acuity and this relationship has previously been reported [32-35]. This finding might reflect a more bullous RD, a pre-operative choroidal detachment or surgical difficulty. These factors are related with retinal glial cell up-regulation and retinal pigment epithelium dispersion, which resulted in a delayed morphologic recovery including that of the photoreceptor reposition of the reattached retina [36,37].

It is thought that the visual results of pseudophakic detachment depend on the pre-operative macula status and approximately 80% of pseudophakic RD typically presents with the macula detached [12-14]. However, 45% of the RDs in this study were macula-on at the time of surgery. We showed that the macula-off RD is not more highly related to poorer outcome, in spite of the worse baseline visual acuity, and that pseudophakic detachment is not present more often with the macula detached. We believe that patients tend to have accurate and frequent eye examinations, which would lead to an earlier diagnosis before there is macular involvement and this is reflected in our final results.

In addition, both vitreous loss at the time of the cataract surgery and Nd:YAG posterior capsulotomy appear to increase the risk of RD when performed after cataract surgery [14]. A defect in the lens capsule may be related to more vitreous involvement and reflect more complex intraocular manipulations. We investigated the defects in the lens capsules and the IOL position, but defects in the lens capsules and the IOL position were not prognostic factors for post-operative outcomes.

Although earlier reports suggested that vitrectomy without SB, the use of perflurocarbon liquids during vitrectomy, and the performance of a drainage retinotomy were associated with the existence of more complex RDs, and consequently poorer functional results [38-40], more recent reports have shown that the surgical outcomes appear to be independent of the surgical procedure that is chosen to resolve the RD [41,42]. In the current study, the type of surgical procedure (vitrectomy or SB) and intra-operative factors were not associated with the post-operative results. Thus, these results suggest that surgical outcomes are independent of the methods or complexity of the surgery.

This study has some limitations. This retrospective series had a relatively small sample size within one retinal center and the surgical procedures that were used were heterogeneous. We did not describe the size of the retinal breaks, vitreous status, or the intra-operative events that occurred during the cataract surgery. A large, multicenter study is needed to evaluate the precise prognostic factors for surgical outcomes in patients with pseudophakic RDs. However, the surgical treatment of all the patients in this study was performed by one experienced retinal specialist. We believe this report is of importance to clinicians who perform surgery for RDs in pseudophakic patients.

Consequently, retinal breaks that were not identified during surgery were related to poor anatomic and functional outcomes. Clinicians should perform careful examinations in the superotemporal quadrant to find retinal breaks when visible breaks are not identified. A poorer pre-operative state, including a worse pre-operative visual acuity, the macula-off RD, a larger area involved in the RD, and multiple breaks was not related to a poorer surgical outcome. Adequate surgery that is performed according to the preference of the surgeon may lead to successful outcomes in patients with pseudophakic RDs.