Purpose
To evaluate refractive changes after lens-sparing vitrectomy for rhegmatogenous retinal detachment (RD).
Design
Retrospective case series.
Methods
A retrospective chart review was conducted in 66 eyes of 66 patients (50.0 ± 9.9 years old) who had undergone lens-sparing vitrectomy for rhegmatogenous RD. Spherical equivalent refractive power was evaluated before and 1, 2, 3, 6, 9, 12, and 15 months after vitrectomy. The relation between refractive changes and several parameters was investigated, such as axial length, presence of preoperative hemorrhage, preoperative spherical equivalent, retinal tear size, logMAR best-corrected visual acuity, number of laser photocoagulations, occurrence of postoperative vitreous hemorrhage, and degree of postoperative inflammatory reaction. Surgical parameters examined included operative time, wide-angle viewing system use, intraoperative adjuvant and gas tamponade use, vitrectomy system gauge, and surgeon.
Results
Significant and continuous myopic shift was observed after vitrectomy throughout the study period. Spherical equivalent was not significantly different between the operated eyes and the fellow control eyes until 3 months after vitrectomy, but the operated eyes were significantly more myopic at 3 months and later postoperatively ( P < .05). Of the 58 eyes finally included (8 patients lost to follow-up), 27 (47%) underwent cataract surgery after vitrectomy. Patients who underwent cataract surgery were significantly older than those who did not ( P < .05); no other examined parameter was significantly different between groups.
Conclusions
A significant myopic progression occurred in eyes after lens-sparing vitrectomy for rhegmatogenous RD. A considerable amount of anisometropia occurred, even in the early postoperative period. Patient age was the only risk factor with the potential to advance the nuclear sclerotic cataract progression after vitrectomy.
Since the development of pars plana vitrectomy using a 17 gauge vitreous cutter in 1970, the surgical techniques employed in vitrectomy have dramatically improved, especially with the recent introduction of transconjunctival, small-gauge, sutureless vitrectomy. Smaller incisions and shorter operation times have reduced the degree of operative invasion and led to faster postoperative recovery. In 1975, Michels and Ryan provided the first report on the occurrence of progressive nuclear sclerosis after vitrectomy in patients with proliferative diabetic retinopathy. Since then, acceleration of nuclear sclerotic cataract and subsequent vision loss have been well recognized as potential adverse surgical outcomes of lens-sparing vitrectomy. Nuclear sclerotic cataracts also frequently affect refraction of the eye, shifting it toward myopic in the majority of cases. To date, most quantitative evaluations of nuclear sclerotic cataracts have been done by evaluating lens opacity with Scheimpflug photography and fluorophotometry. Wong and associates reported the progression of lens opacity and scattering in patients with macular hole using Scheimpflug photography. Ogura and associates used fluorophotometry and found that lens autofluorescence in vitrectomized eyes was significantly higher than in the contralateral control eyes. To the best of our knowledge, there is only 1 report of myopic shift after pars plana vitrectomy. However, that study did not involve longitudinal evaluation.
The purpose of this study was to evaluate the time course of refractive changes after lens-sparing vitrectomy for rhegmatogenous retinal detachment (RD).
Methods
This retrospective study included 64 vitrectomized eyes of 64 patients (50 male and 14 female) averaging 50.0 ± 9.9 years of age (mean ± standard deviation) (range: 13–65 years). All subjects had undergone lens-sparing vitrectomy for rhegmatogenous RD at Tsukuba University Hospital between July 1, 2008 and April 30, 2013. To be included in the study, patients had to have a primary rhegmatogenous RD and be phakic. None of the patients had a visually significant cataract at the time of surgery. Patients were excluded from analyses if they had a history of ocular surgery, or if ocular disease was present that could affect refraction and/or visual function. This study was approved retrospectively by the Institutional Review Board at the Tsukuba University Hospital, and study conduct adhered to the tenets of the Declaration of Helsinki.
All surgeries were performed under local trans-Tenon retrobulbar anesthesia by 1 of 2 surgeons (F.O., Y.O.). One surgeon (F.O.) operated on 47 eyes using a 23 gauge vitrectomy system, and the other (Y.O.) operated on 19 eyes using a 25 gauge vitrectomy system. All surgical procedures were performed using the Accurus microincision vitrectomy system (Alcon Laboratories, Fort Worth, Texas, USA). Contact lenses (truncated contact lens; HOYA Corp, Tokyo, Japan) and a wide-angle viewing system (Resight; Carl Zeiss Meditec AG, Jena, Germany) were used to facilitate posterior visualization during surgery. The total gas-fluid exchange used air or 20% sulfur hexafluoride (SF6). To secure vitreous traction release around the breaks and prevent vitreous incarceration of the sclerotomy wound, triamcinolone acetonide was used as needed. All patients were kept in the prone position for a few days after surgery.
Preoperative and postoperative refractive errors were measured in all eyes using an automatic refractometer (RK-2; Canon Inc, Tokyo, Japan). Axial length was evaluated using noncontact optical biometry (IOLMaster; Carl Zeiss Meditec AG) before primary vitrectomy.
Each clinical parameter was evaluated at baseline (before surgery) and 1, 2, 3, 6, 9, 12, and 15 months after lens-sparing vitrectomy. Data were analyzed using repeated-measures analysis of variance (ANOVA) to assess the time course of changes. If significant differences were observed, a Dunnett test was performed. Data obtained from the operative and contralateral eyes were also compared, and differences were tested for statistical significance using Student t test. A single linear regression was used to calculate regression coefficients of refractive power, which was chosen as an outcome variable. Risk factors for cataract progression were assessed by comparing patient demographic characteristics and intraoperative and postoperative parameters in patients who underwent cataract surgery in the follow-up period and those who did not. Statistical significance was tested using Student t test or the χ 2 test, as appropriate. Correlations between refractive power progression rates after lens-sparing vitrectomy and patient age, axial length, preoperative refractive power, circumferential retinal tear dimension, operation time, and the photocoagulation number were analyzed by Spearman rank correlation. Refractive progression rate was defined as the mean refractive change per month between the baseline and final visits. When refractive power could not be measured before lens-sparing vitrectomy because of macular detachment or vitreous hemorrhage, the refraction obtained at the first postvitrectomy visit was defined as the baseline refraction. Statistical significance was defined as P < .05. All statistical analyses were performed with StatView statistical software (version 5.0; SAS Institute, Cary, North Carolina, USA).
Results
Patient clinical characteristics, intraoperative parameters, and postoperative complications are summarized in Table 1 . No patient had a rhegmatogenous RD secondary to a globe contusion, or had proliferative vitreoretinopathy. The retina was successfully reattached with the initial operation in all eyes, and no serious intraoperative or postoperative complications (eg, subretinal hemorrhage, elevated intraocular pressure for >7 days, choroidal detachment) were observed. A gas-fluid exchange was performed in 64 of 66 patients and the gas (air or 20% SF6) was selected according to ocular status. Two patients did not receive gas-fluid exchange owing to the limited area of their retinal detachment.
| Age (y) | 50.0 ± 9.9 |
| Sex (male/female) | 50/14 |
| Ocular side (left/right) | 25/39 |
| Preoperative vitreous hemorrhage (yes/no) | 12/52 |
| Axial length (mm) (24 eyes) | 26.0 ± 1.9 |
| Preoperative spherical equivalent (diopter) (38 eyes) | −5.68 ± 3.71 |
| Circumferential dimension of retinal tears (degrees) | 62.3 ± 33.5 |
| Macular status (on/off) | 27/37 |
| Operation time (min) | 51.3 ± 17.4 |
| Gauge of vitrectomy system (23 gauge/25 gauge) | 47/17 |
| Use of wide-angle viewing system (yes/no) | 9/55 |
| Laser photocoagulation number (shots) | 783 ± 454 |
| Use of gas tamponade (none/air/20% SF6) | 2/13/49 |
| Use of triamcinolone acetonide for vitreous visualization (yes/no) | 21/42 |
| Occurrence of vitreous hemorrhage (yes/no) | 2/62 |
| Occurrence of fibrin reaction (yes/no) | 2/62 |
| BCVA following vitrectomy (logMAR) (Snellen equivalent) | 0.23 ± 0.33 (20/34) |
After lens-sparing vitrectomy, patients were followed up for an average of 11.6 ± 11.8 months. Following vitrectomy, 6 of 64 patients did not return to our clinic and were lost to follow-up. Of the remaining 58 eyes, 27 (46.6%) underwent cataract surgery during the study period. Cataract type was nuclear sclerosis in every case, and there was no case of cortical or posterior subcapsular cataract. All cataract surgeries were successful and uneventful. Neither intraoperative (eg, lens capsule rupture) nor postoperative (eg, zonule formation, persistent ocular hypertension) complications were observed during the follow-up period.
Table 2 demonstrates the time course of refractive changes after lens-sparing vitrectomy for rhegmatogenous RD. The operative eyes showed a progressive myopic shift ( Figure 1 ), and the refractions 6 months and later after vitrectomy were significantly different from baseline ( P < .05, Dunnett test). The refractions in the contralateral eyes were stable throughout the study period ( Figure 1 ). There was no significant difference in postoperative spherical equivalent between the operative and contralateral eyes until 3 months after surgery, but the operative eyes were more myopic than the contralateral eyes 3 months or later after lens-sparing vitrectomy (all P < .05, Student t test). A significant negative correlation was found between time after vitrectomy and spherical equivalent in the operative eyes (R 2 = 0.216, P < .001), but not in the contralateral eyes (R 2 = 0.016, P = .78).
| Operated Eyes (Diopter) | Average Deviation From Baseline (Diopter) | Contralateral Eyes (Diopter) | Degree of Anisometropia (Diopter) | P Value | |
|---|---|---|---|---|---|
| Baseline | −5.78 ± 3.69 | −6.14 ± 3.44 | 0.36 | .49 | |
| 1 month | −6.89 ± 4.17 | −1.11 | −5.87 ± 3.42 | 1.02 | .18 |
| 2 months | −7.44 ± 2.92 | −1.66 | −6.29 ± 2.38 | 1.15 | .44 |
| 3 months | −7.67 ± 4.40 | −1.89 | −5.65 ± 3.26 | 2.02 | .05 b |
| 6 months | −8.54 ± 4.21 a | −2.76 | −5.42 ± 3.44 | 3.12 | .005 b |
| 9 months | −9.43 ± 4.17 a | −3.65 | −6.00 ± 2.82 | 3.43 | .04 b |
| 12 months | −9.81 ± 4.58 a | −4.03 | −6.25 ± 2.68 | 3.56 | .05 b |
| 15 months | −10.75 ± 4.65 a | −4.97 | −5.51 ± 2.79 | 5.24 | .03 b |
a Statistical difference from baseline (Dunnett test).
b Statistical difference between operative and contralateral eyes (Student t test).
When comparison of baseline clinical characteristics was performed between patients who underwent cataract surgery in the follow-up period (27 eyes, 46.6%) and patients who did not (31 eyes, 53.4%), patients who underwent cataract surgery were significantly older than patients who did not (50.9 ± 15.1 years and 49.7 ± 18.9 years, respectively, P < .001). Differences between these groups relating to sex, ocular side, axial length, preoperative spherical equivalent, and circumferential dimension of retinal tears were not significant ( P = .52, P = .44, P = .41, P = .45, and P = .26, respectively). Furthermore, when comparisons of intraoperative parameters including operative time, gauge of vitrectomy system, use of wide-angle viewing system, laser photocoagulation number, use of gas tamponade, and best-corrected visual acuity (BCVA) were performed between these 2 groups, significant differences were not observed for any of them ( P = .79, P = .08, P = .11, P = .16, P = .58, and P = .19, respectively).
The progression rate of the refractive spherical equivalent following vitrectomy was significantly correlated with patient age ( P < .01; Figure 2 ). However, no significant relationships were found between the spherical equivalent progression rate and axial length ( P = .14), preoperative spherical equivalent ( P = .69), retinal tear circumference ( P = .53), operation time ( P = .37), or the number of photocoagulation spots ( P = .61).
