Abstract
AIM
To identify risk factors for the development of posterior synechiae of the iris (PSI) after 23-gauge phacovitrectomy.
METHODS
A retrospective chart review was performed in consecutive Asian patients treated with 23-gauge phacovitrectomy with a 3-piece intraocular lens (IOL) or a single-piece 4 haptics IOL.
RESULTS
A total of 263 eyes from 242 patients were included in the study. Postoperative PSI was identified in 16 (6.1%) eyes. In multivariate analysis, C3F8 gas tamponade, oil tamponade, and long operation time were significantly associated with PSI formation. There was no difference in the incidence of PSI between the groups using two different types of IOL (P=0.779).
CONLUSION
C3F8 gas or oil tamponade and long operation time increased the incidence of PSI after 23-gauge phacovitrectomy. The single-piece 4 haptics IOL, in lieu of a 3-piece IOL, may be inserted into the capsular bag with a comparable incidence of PSI.
Keywords: complication, intraocular lens, phacoemulsification, vitrectomy
INTRODUCTION
Phacoemulsification during pars plana vitrectomy (PPV) is a commonly accepted surgical procedure for patients with vitreoretinal problems and cataracts; however, one of the complications of this procedure is formation of posterior synechiae of the iris (PSI), which may lead to a decrease in visual quality, difficulties in peripheral retinal examination due to the impairment of pupillary movement, and rarely secondary angle-closure glaucoma[1]. In our previous study on eyes treated with conventional 20-gauge PPV and phacoemulsification, preoperative PSI, C3F8 gas tamponade, postoperative anterior chamber fibrin deposition, and a single-piece intraocular lens (IOL) were shown to increase the incidence of postoperative PSI[2]. PSI developed in 18.9% of eyes in the 3-piece IOL group and 34.0% of eyes in the single-piece IOL group (P=0.041)[2]. Recently, the operation environment has changed radically with the development of new surgical instruments. Transconjunctival sutureless 23-gauge PPV has replaced conventional 20-gauge PPV for most vitrectomy cases. It has been reported that the 23-gauge system provides reduced surgical operation time and postoperative inflammation compared to the 20-gauge system[3]–[7]. Therefore, the aspect of postoperative PSI formation would have been different when using 23-gauge system. In addition, the 3-piece IOL is somewhat difficult to use when it is inserted into the bag. Because of its ease of insertion and its stability in the capsular bag, a single-piece 4 haptics IOL has gained popularity when cataract surgery is combined with PPV[8]–[10].
The primary purpose of this study was to evaluate risk factors for PSI in combined transconjunctival sutureless 23-gauge PPV and clear cornea phacoemulsification with an in-the-bag IOL insertion (23-gauge phacovitrectomy). In addition, we compared the incidence of postoperative PSI between the groups with two different types of IOL (3-piece IOL vs single-piece 4 haptics IOL).
SUBJECTS AND METHODS
Approval was obtained from the Korea University Ansan Hospital Institutional Review Board, Ansan, Korea. All research and data collection followed the tenets of the Declaration of Helsinki. The procedures were fully explained to each patient, and each provided written informed consent.
Subjects
Case selection and data collection
We retrospectively reviewed the medical records of consecutive patients treated with phacovitrectomy by one surgeon (Kim SW) at the Korea University Ansan Hospital, Ansan, Korea, between September 2008 and August 2012. Exclusion criteria included a follow-up period of less than 3mo, intraoperative capsular radial tears, posterior capsular injury, zonular dehiscence leading to IOL decentering, and surgery for ocular trauma (e.g. corneal or scleral laceration, traumatic retinal detachment, traumatic cataract, or traumatic lens dislocation).
Preoperative data, including age, gender, a history of diabetes or uveitis, presence of diabetic retinopathy (DR), presence of retinal new vessels, presence of preoperative PSI, presence of neovascularization of the iris (NVI), anterior chamber depth (ACD), axial length (AL), and major indications for vitrectomy were collected. Intraoperative data, including IOL type, number of endolaser photocoagulation, type of tamponade, and the total operation time of phacovitrectomy (min), and postoperative data, including history of postoperative anterior chamber fibrin formation, were also obtained.
Definition of posterior synechiae of the iris
PSI was detected by slit-lamp biomicroscopy. A PSI was defined as being present when it was recognized over 30 degrees of the pupillary margin after pupillary dilation with tropicamide and 0.5% phenylephrine hydrochloride (Mydrin®-P, Santen Pharmaceutical Co., Ltd, Osaka, Japan) at two or more examinations conducted during the initial three postoperative months[11]. Transient synechiae or those released by pupillary dilation were excluded.
Methods
All operations were conducted under local anesthesia. In all cases, 23-gauge trocars and cannulas, an Associate DORC posterior vitrectomy system (DORC, Inc., Zuidland, Netherlands), a Stellaris phacoemulsification system (Bausch and Lomb, Rochester, NY, USA), and a binocular indirect ophthalmo-microscope system (Oculus, Wetzlar, Germany) were used. All eyes underwent implantation of a 3-piece hydrophobic acrylic IOL (Tecnis™ Acrylic IOL ZA9003; Abbott Medical Optics, Inc. Santa Ana, CA, USA) or a single-piece hydrophilic 4 haptics IOL (Akreos MI60; Bausch & Lomb, Rochester, NY, USA). In more recent cases, single-piece 4 haptics IOLs were used.
At first, transconjunctival cannulas were inserted in a two-step procedure through the pars plana in the inferotemporal, superonasal, and superotemporal quadrants. The cannulas were temporarily closed with plugs, and the infusion was closed while the temporal corneal phacoemulsification was performed. A 2.75-mm temporal clear corneal incision was made regardless of IOL type. After a 5.0 to 5.5 mm-sized continuous curvilinear capsulorhexis was created using a bent 27-gauge needle and capsulorhexis forceps, hydrodissection and hydrodelineation of the lens were performed. The lens was removed using a phaco-handpiece. After the IOL was inserted into the capsular bag, the corneal incision was secured using a single 10-0 nylon suture and transconjunctival sutureless 23-gauge PPV was then performed. Triamcinolone acetonide was injected intraoperatively in all cases after core and peripheral vitrectomy to visualize the remaining peripheral vitreous and vitreoschisis or epiretinal membrane (ERM) at the posterior pole. Then, triamcinolone acetonide in the vitreous cavity was removed as much as possible. Complete removal of the fibrovascular tissue was attempted in every case in which fibrovascular proliferation was noted in eyes with proliferative DR or retinal detachment. The ERM or internal limiting membrane was peeled in eyes with macular pucker, macular holes, or macular edema. Endolaser photocoagulation using an argon laser was added when necessary. Bleeding during vitrectomy was controlled by diathermy or by increasing the infusion pressure. If oil or gas tamponade was required, an endotamponade was performed at the end of the surgery with 14% C3F8 gas or 1300 CS silicone oil. Subconjunctival injection of steroids or antibiotics was not performed at the end of the surgery.
Early postoperative management
All patients were discharged from the hospital one or two days after surgery. Patients with gas tamponade were instructed to keep in the face-down position for one month; however, this duration was customized according to the patient's general and ocular condition. The routine postoperative regimen included instillation of a 1% prednisolone acetate ophthalmic solution and moxifloxacin ophthalmic solution every two hours during the day. Dexamethasone ointment was applied before sleeping. From postoperative 10d, patients were treated with 1% prednisolone acetate ophthalmic solution four times a day until inflammatory cells disappeared from the anterior chamber.
Statistical Analysis
Statistical analysis was performed using SPSS version 20.0 (SPSS Inc., Chicago, IL, USA). All statistics were two-tailed, and P-values less than 0.05 were considered significant. The Kolmogorov-Smirnov test was used to verify the normality of the distribution of continuous variables. For comparison between groups, the Chi-square analysis or Fisher's exact test was used for categorical variables and the t-test or Mann-Whitney U test was used for continuous variables. To identify risk factors for postoperative PSI, multivariate logistic regression analysis was conducted. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated.
RESULTS
During the study period, 322 eyes underwent phacovitrectomy using the 23-gauge TSV system. Of these, 59 eyes were excluded because of early loss to follow-up (28 eyes), capsular or zonular problems (25 eyes), and ocular trauma (6 eyes). Finally, a total of 263 eyes from 242 Asian patients were included in the study. There were no cases in which iris retractor was used or radial iris sphincter incision was performed to stretch the small pupil. The mean age of included patients was 61.3±13.07y. The two major indications for surgery were ERM and vitreous hemorrhage associated with DR, followed by rhegmatogenous retinal detachment (RRD) and retinal traction associated with DR (Table 1).
Table 1. Surgical indications and formation of postoperative posterior synechiae of the iris.
| Indication | Postoperative posterior synechiae |
|
| No (n=247) | Yes (n=16) (%)1 | |
| Epiretinal membrane | 102 | 3 (2.9) |
| VH associated with DR | 61 | 1 (1.6) |
| Rhegmatogenous retinal detachment | 20 | 8 (28.6) |
| Retinal traction associated with DR | 15 | 3 (16.7) |
| Refractory diabetic macular edema | 14 | 1 (6.7) |
| VH associated with RVO | 13 | 0 (0) |
| Macular hole | 10 | 0 (0) |
| Vitreous opacity associated with uveitis | 6 | 0 (0) |
| VH associated with other diseases | 4 | 0 (0) |
| Retinal traction associated with RVO | 2 | 0 (0) |
VH: Vitreous hemorrhage; DR: Diabetic retinopathy; RVO: Retinal vein occlusion. 1Incidence of postoperative posterior synechiae of the iris in total cases of stated indication.
After phacovitrectomy, PSI developed in 16 (6.1%) of total 263 eyes. There were no significant differences between the groups with and without postoperative PSI in mean age, gender, history of diabetes, laterality of the eye, mean AL, mean ACD, presence of DR, presence of retinal new vessels, presence of preoperative PSI, presence of NVI, IOL type, history of endolaser photocoagulation, or history of postoperative anterior chamber fibrin formation (Table 2). However, number of endolaser photocoagulation, history of gas or oil tamponade, and operation time were significantly different between the two groups (P=0.035, P<0.001, and P<0.001 respectively). Six (2.3%) of total 263 eyes had a history of uveitis and all of these eyes underwent vitrectomy due to vitreous opacity. Postoperative PSI did not develop in any of these eyes with a history of uveitis.
Table 2. Comparison of characteristics between the groups with and without postoperative posterior synechiae of the iris.
| Variable | Postoperative posterior synechiae |
P | |
| No (n=247) | Yes (n=16) (%)1 | ||
| Age (a) | 61.6±12.9 | 56.8±15.1 | 0.258 |
| Gender (n) | 0.731 | ||
| M | 128 | 9 (6.6) | |
| F | 119 | 7 (5.6) | |
| History of diabetes (n) | 0.491 | ||
| No | 118 | 9 (7.1) | |
| Yes | 129 | 7 (5.1) | |
| Laterality (n) | 0.063 | ||
| OD | 126 | 12 (8.7) | |
| OS | 121 | 4 (3.2) | |
| Axial length (mm) | 23.5±1.3 | 23.8±1.1 | 0.050 |
| Anterior chamber depth (mm) | 2.91±0.55 | 2.72±0.99 | 0.709 |
| Diabetic retinopathy (n) | 0.481 | ||
| No | 132 | 10 (7.0) | |
| Yes | 115 | 6 (5.0) | |
| Retinal new vessels (n) | 0.876 | ||
| No | 134 | 9 (6.3) | |
| Yes | 113 | 7 (5.8) | |
| Preoperative posterior synechiae (n) | 20.117 | ||
| No | 239 | 14 (5.5) | |
| Yes | 8 | 2 (20.0) | |
| Preoperative NVI (n) | 20.223 | ||
| No | 244 | 15 (5.8) | |
| Yes | 3 | 1 (25) | |
| IOL type (n) | 0.779 | ||
| 3-piece | 130 | 9 (6.5) | |
| 4 haptics | 117 | 7 (5.6) | |
| Intraoperative endolaser photocoagulation (n) | 20.768 | ||
| No | 63 | 3 (4.5) | |
| Yes | 184 | 13 (6.6) | |
| Number of endolaser photocoagulation (n) | 457±465 | 688±471 | 0.035 |
| Gas or oil tamponade (n) | <0.001 | ||
| No | 173 | 1 (0.6) | |
| Yes | 74 | 15 (16.9) | |
| Operation time (min) | 70.9±32.5 | 113.1±45.4 | <0.001 |
| Postoperative AC fibrin formation (n) | 20.201 | ||
| No | 224 | 13 (5.5) | |
| Yes | 23 | 3 (11.5) | |
NVI: Neovascularization of the iris; IOL: Intraocular lens; AC: Anterior chamber. 1Incidence of postoperative posterior synechiae of the iris in total cases of the stated indication. The Chi-square analysis or 2Fisher's exact test was used for categorical variables and the Mann-Whitney U test was used for continuous variables for comparison between the two groups.
In a subgroup analysis of the eyes that underwent gas or oil tamponade (n=89), postoperative PSI developed in 15 (16.9%) cases. Only operation time was significantly different between the groups with and without postoperative PSI (P=0.018, Table 3). In a subgroup analysis of the eyes whose underlying surgical indication was retinal traction associated with DR or RRD (n=46), postoperative PSI developed in 11 (23.9%) cases and all cases underwent gas or oil tamponade (Table 4). Mean ACD was shallower in the group with postoperative PSI (P=0.020).
Table 3. Subgroup analysis in the eyes that underwent gas or oil tamponade.
| Variable | Postoperative posterior synechiae |
P | |
| No (n=74) | Yes (n=15) (%)1 | ||
| Age (a) | 60.1±13.5 | 56.7±15.6 | 0.493 |
| Gender (n) | 0.523 | ||
| M | 46 | 8 (14.8) | |
| F | 28 | 7 (20.0) | |
| History of diabetes (n) | 0.423 | ||
| No | 36 | 9 (20.0) | |
| Yes | 38 | 6 (13.6) | |
| Laterality (n) | 0.169 | ||
| OD | 40 | 11 (21.6) | |
| OS | 34 | 4 (10.5) | |
| Axial length (mm) | 23.4±1.0 | 23.8±1.1 | 0.118 |
| Anterior chamber depth (mm) | 2.92±0.53 | 2.62±0.94 | 0.313 |
| Diabetic retinopathy (n) | 0.369 | ||
| No | 40 | 10 (20.0) | |
| Yes | 34 | 5 (12.8) | |
| Retinal new vessels (n) | 0.673 | ||
| No | 40 | 9 (18.4) | |
| Yes | 34 | 6 (15.0) | |
| Preoperative posterior synechiae (n) | 20.336 | ||
| No | 69 | 13 (15.9) | |
| Yes | 5 | 2 (28.6) | |
| Preoperative NVI (n) | 20.999 | ||
| No | 73 | 15 (17.0) | |
| Yes | 1 | 0 (0) | |
| IOL type (n) | 0.321 | ||
| 3-piece | 34 | 9 (20.9) | |
| 4 haptics | 40 | 6 (13.0) | |
| Intraoperative endolaser photocoagulation (n) | 20.698 | ||
| No | 11 | 3 (21.4) | |
| Yes | 63 | 12 (16.0) | |
| Number of endolaser photocoagulation (n) | 674±594 | 712±477 | 0.591 |
| Tamponade type (n) | 20.999 | ||
| C3F8 gas | 65 | 13 (16.7) | |
| Silicone oil | 9 | 2 (18.2) | |
| Operation time (min) | 86.8±39.6 | 114.1±46.8 | 0.018 |
| Postoperative AC fibrin formation (n) | 20.999 | ||
| No | 62 | 12 (16.2) | |
| Yes | 12 | 3 (20.0) | |
NVI: Neovascularization of the iris; IOL: Intraocular lens; AC: Anterior chamber. 1Incidence of postoperative posterior synechiae of the iris in total cases of the stated indication. The Chi-square analysis or 2Fisher's exact test was used for categorical variables and the Mann-Whitney U test was used for continuous variables for comparison between the two groups.
Table 4. Subgroup analysis in the eyes whose underlying surgical indication was retinal traction associated with diabetic retinopathy or rhegmatogenous retinal detachment.
| Variable | Postoperative posterior synechiae |
P | |
| No (n=35) | Yes (n=11) (%)1 | ||
| Age (a) | 51.3±15.3 | 58.6±15.9 | 0.122 |
| Gender (n) | 20.494 | ||
| M | 21 | 5 (19.2) | |
| F | 14 | 6 (30.0) | |
| History of diabetes (n) | 0.478 | ||
| No | 18 | 7 (28.0) | |
| Yes | 17 | 4 (19.0) | |
| Laterality (n) | 20.732 | ||
| OD | 19 | 7 (26.9) | |
| OS | 16 | 4 (20.0) | |
| Axial length (mm) | 23.3±1.4 | 23.8±1.3 | 0.252 |
| Anterior chamber depth (mm) | 3.05±0.43 | 2.40±1.00 | 0.020 |
| Diabetic retinopathy (n) | 20.486 | ||
| No | 20 | 8 (28.6) | |
| Yes | 15 | 3 (16.7) | |
| Retinal new vessels (n) | 20.732 | ||
| No | 19 | 7 (26.9) | |
| Yes | 16 | 4 (20.0) | |
| Preoperative posterior synechiae (n) | 20.999 | ||
| No | 31 | 10 (24.4) | |
| Yes | 4 | 1 (20.0) | |
| Preoperative NVI (n) | |||
| No | 35 | 11 (23.9) | |
| Yes | 0 | 0 (0) | |
| IOL type (n) | 20.169 | ||
| 3-piece | 13 | 7 (35.0) | |
| 4 haptics | 22 | 4 (15.4) | |
| Intraoperative endolaser photocoagulation (n) | 20.999 | ||
| No | 4 | 1 (20.0) | |
| Yes | 31 | 10 (24.4) | |
| Number of endolaser photocoagulation (n) | 862±633 | 845±439 | 0.877 |
| Gas or oil tamponade (n) | 20.169 | ||
| No | 8 | 0 (0) | |
| Yes | 27 | 11 (28.9) | |
| Operation time (min) | 107.1±43.5 | 121.1±46.2 | 0.309 |
| Postoperative AC fibrin formation (n) | 20.999 | ||
| No | 29 | 9 (23.7) | |
| Yes | 6 | 2 (25.0) | |
NVI: Neovascularization of the iris; IOL: Intraocular lens; AC: Anterior chamber. 1Incidence of postoperative posterior synechiae of the iris in total cases of the stated indication. The Chi-square analysis or 2Fisher's exact test was used for categorical variables and the Mann-Whitney U test was used for continuous variables for comparison between the two groups.
Table 5 shows a comparison of characteristics between the groups with two different types of IOL. Eyes with DR or retinal new vessels were more common in the group using 4 haptics IOL (P=0.014 and P=0.002 respectively).
Table 5. Comparison of characteristics between the groups with 3-piece intraocular lenses (IOLs) and single-piece 4 haptics IOLs.
| Variable | IOL type |
P | |
| 3-piece (n=139) | 4 haptics (n=124) | ||
| Age (a) | 62.4±12.2 | 60.1±13.9 | 0.150 |
| Gender (n) | 0.167 | ||
| M | 78 | 59 | |
| F | 61 | 65 | |
| History of diabetes (n) | 0.181 | ||
| No | 72 | 54 | |
| Yes | 67 | 70 | |
| Laterality (n) | 0.792 | ||
| OD | 74 | 64 | |
| OS | 65 | 60 | |
| Axial length (mm) | 23.6±1.3 | 23.3±1.3 | 0.072 |
| Anterior chamber depth (mm) | 2.83±0.66 | 2.97±0.49 | 0.081 |
| Diabetic retinopathy (n) | 0.014 | ||
| No | 85 | 57 | |
| Yes | 54 (38.8%) | 67 (54.0%) | |
| Retinal new vessels (n) | 0.002 | ||
| No | 88 | 55 | |
| Yes | 51 (36.7%) | 69 (55.6%) | |
| Preoperative posterior synechiae (n) | 20.524 | ||
| No | 135 | 118 | |
| Yes | 4 | 6 | |
| Preoperative NVI (n) | 20.346 | ||
| No | 138 | 121 | |
| Yes | 1 | 3 | |
| Intraoperative endolaser photocoagulation (n) | 0.750 | ||
| No | 36 | 30 | |
| Yes | 103 | 94 | |
| Number of endolaser photocoagulation (n) | 465±453 | 479±486 | 0.807 |
| Gas or oil tamponade (n) | 0.292 | ||
| No | 96 | 78 | |
| Yes | 43 | 46 | |
| Operation time (min) | 75.8±35.8 | 70.8±33.6 | 0.256 |
| Postoperative AC fibrin formation (n) | 0.257 | ||
| No | 128 | 109 | |
| Yes | 11 | 15 | |
IOL: Intraocular lens; NVI: Neovascularization of the iris; AC: Anterior chamber. The Chi-square analysis or 2Fisher's exact test was used for categorical variables and the t-test was used for continuous variables for comparison between the two groups.
Multivariate logistic regression analysis was performed using several variables including age, gender, presence of DR, presence of retinal new vessels, AL, ACD, IOL type, number of endolaser photocoagulation, tamponade type, and operation time. C3F8 gas tamponade, oil tamponade, and long operation time were significantly associated with postoperative PSI formation (Table 6). The OR of C3F8 gas tamponade for predicting postoperative PSI formation was 29.7 (95%CI=3.28 to 269.07), the OR of oil tamponade was 27.0 (95%CI=1.28 to 567.61), and the OR of operating time (min) was 1.02 (95%CI=1.00 to 1.04).
Table 6. Multivariate logistic regression results for predicting postoperative posterior synechiae formation after 23-gauge phacovitrectomy.
| Variable | Odds ratio | 95%CI | P |
| Tamponade type | |||
| No tamponade | ref. | ||
| C3F8 gas | 29.7 | 3.28 to 269.07 | 0.003 |
| Silicone-oil | 27.0 | 1.28 to 567.61 | 0.034 |
| Operation time (min) | 1.02 | 1.00 to 1.04 | 0.029 |
| Age (a) | 0.97 | 0.91 to 1.03 | 0.279 |
| Gender (M) | 0.41 | 0.10 to 1.59 | 0.195 |
| Presence of diabetic retinopathy | 0.47 | 0.05 to 4.24 | 0.501 |
| Presence of retinal new vessels | 1.86 | 0.22 to 15.49 | 0.564 |
| Axial length (mm) | 1.19 | 0.70 to 2.03 | 0.523 |
| Anterior chamber depth (mm) | 0.49 | 0.19 to 1.23 | 0.129 |
| IOL type (4 haptics) | 0.81 | 0.20 to 3.21 | 0.763 |
| Number of endolaser photocoagulation (n) | 1.00 | 0.99 to 1.00 | 0.543 |
IOL: Intraocular lens.
DISCUSSION
PSI are formed from fibrous metaplasia of the inflamed iris tissue caused by early postoperative iritis[12],[13]. In the present study, the incidence of PSI was 6.1%. In our previous study, the incidence of PSI was 23.5%, and other studies reported a frequency of PSI formation after phacovitrectomy between 7% and 30%[2],[11],[14]–[18]. It is difficult to compare the incidence of PSI in the present study with the results of other studies because of the different study design, subject characteristics, and surgical techniques. Patient demographic characteristics could be an important explanation for the low incidence of postoperative PSI in this study. The most relevant surgical indications in this study were ERM and DR-associated vitreous hemorrhage, which comprised 63.5% of all cases. Generally, these indications are not considered challenging from a surgical point of view, resulting in less surgical trauma and inflammation within a relatively short operation time. No tamponade with a lack of major surgical complications is the usual surgical scenario in these cases. In addition, the use of the 23-gauge system, which provides reduced surgical operation time and less postoperative inflammation compared to the 20-gauge system, may have affected the relatively low incidence in this study[3]–[5].
Another reason for the low incidence of PSI may be the type of IOL used in the present study. Previously, we inserted a 3-piece hydrophobic acrylic IOL (Acrysof MA60BM; Alcon labs, Fort Worth, TX, USA) or a single-piece hydrophobic acrylic IOL (Acrysof SA60AT; Alcon labs, Fort Worth, TX, USA). PSI developed in 18.9% of eyes in the 3-piece IOL group and 34.0% of eyes in the single-piece IOL group in that study[2]. A single-piece IOL was a risk factor for PSI formation. In eyes with gas or oil tamponade, the incidence of PSI was 32.6% (15/46) in the 3-piece IOL group and 52.9% (9/17) in the single-piece IOL group[2]. In the present study, we used a single-piece 4 haptics IOL or a 3-piece IOL. The single-piece 4 haptics IOL seemed to be comparable to the 3-piece IOL in terms of postoperative PSI formation. PSI developed in 5.6% of eyes in the single-piece 4 haptics IOL group and 6.5% of eyes in the 3-piece IOL group. Even in the subgroup analysis of eyes with gas or oil tamponade, there was no difference in PSI formation between the single-piece 4 haptics IOL group and the 3-piece IOL group [13.0% (6/46) vs 20.9% (9/43), P=0.321].
These comparable results of single-piece 4 haptics IOL with the 3-piece IOL in PSI formation may be due to its mechanical properties. The most prominent advantage of the single-piece 4 haptics IOL over the 3-piece IOL was its ease of insertion and the management of the IOL in the capsular bag, probably resulting from its flexibility and thin optic portion. In spite of its flexibility, the four legs of the single-piece 4 haptics IOL allowed it to remain stable in the capsular bag during infusion and aspiration after IOL insertion and during the vitrectomy procedure. Besides its four haptic design, the single-piece 4 haptics IOL has a 10-degree haptic angle and a wide area at the optic haptic junction, which resisted the buoyancy of the expanding gas and helped maintain its optic portion behind the iris plane level[8]. Furthermore, beyond its mechanical properties, the hydrophilic characteristics of the single-piece 4 haptics IOL may be another reason for the low incidence of PSI in the single-piece 4 haptics IOL group. Amon and Menapace[19] demonstrated that hydrophilic surfaces showed cellular reactions in a lower percentage of cases compared with hydrophobic surfaces.
Known risk factors of PSI are preoperative PSI, gas or oil tamponade, single-piece IOL implantation, postoperative anterior chamber fibrin deposition, the existence of proliferative DR, and the amount of photocoagulation[2],[11],[14],[19]. In the present study, C3F8 gas or oil tamponade and long operation time were significantly associated with postoperative PSI formation. Long operation time may be associated with increase of postoperative inflammation and consequently, increase of PSI formation. As already noted by several studies, gas or oil tamponade may increase PSI formation due to the tighter apposition of the IOL to the posterior iris surface[2],[11],[15],[20],[21]. In previous our study, uveitis history, preoperative PSI, and postoperative anterior chamber fibrin deposition were significantly higher in the group with postoperative PSI[2]. But these variables were not significant in this study. This may be due to low incidence of postoperative PSI in this study using of the 23-gauge system. In subgroup analysis, preoperative ACD was shallower in the group with postoperative PSI (Table 4). Although preoperative ACD was not significant for predicting postoperative PSI formation in multivariate analysis, the shallower preoperative ACD may be associated with the tighter apposition of the IOL to the posterior iris surface.
There are certain limitations to this study, including its retrospective design and the heterogeneity of surgical indications for vitrectomy. PSI may be associated with intraoperative unusual findings in the anterior chamber such as prolapsed iris, iris damage, or bleedings. However, retrospective design of this study prevented further analysis. The surgical indications for vitrectomy were so diverse that it was difficult to categorize them into small subgroups. Furthermore, even for the same indication, the severity of disease was different and very hard to present in a single scale. Therefore, we could not directly evaluate the effect of each disease on PSI formation; however, we could assess the severity and the effect of the disease indirectly by considering various surgical factors. In addition, the surgical indications for vitrectomy might have some degree of interaction with tamponade type, which is one of the risk factors for PSI. For example, macular holes or RRD surgery always require an endotamponade, which might increase the incidence of PSI in these diseases.
In conclusion, C3F8 gas or oil tamponade and long operation time during 23-gauge phacovitrectomy increased the possibility of postoperative PSI formation. The single-piece 4 haptics IOL, in lieu of a 3-piece IOL, may be inserted into the capsular bag to achieve a comparable incidence of postoperative PSI.
Acknowledgments
Conflict of Interests: Oh JH, None; Na J, None; Kim SW, None; Oh J, None; Huh K, None.
REFERENCES
- 1.Makhzoum O, Crosby NJ, Hero M. Secondary angle-closure glaucoma due to posterior synechiae formation following phacovitrectomy. Int Ophthalmol. 2011;31(6):481–482. doi: 10.1007/s10792-011-9475-3. [DOI] [PubMed] [Google Scholar]
- 2.Kim SW, Oh J, Song JS, Kim YY, Oh IK, Huh K. Risk factors of iris posterior synechia formation after phacovitrectomy with three-piece acrylic IOL or single-piece acrylic IOL. Ophthalmologica. 2009;223(4):222–227. doi: 10.1159/000203366. [DOI] [PubMed] [Google Scholar]
- 3.Park DH, Shin JP, Kim SY. Comparison of clinical outcomes between 23-gauge and 20-gauge vitrectomy in patients with proliferative diabetic retinopathy. Retina. 2010;30(10):1662–1670. doi: 10.1097/IAE.0b013e3181d95261. [DOI] [PubMed] [Google Scholar]
- 4.Misra A, Ho-Yen G, Burton RL. 23-gauge sutureless vitrectomy and 20-gauge vitrectomy: a case series comparison. Eye (Lond) 2009;23(5):1187–1191. doi: 10.1038/eye.2008.175. [DOI] [PubMed] [Google Scholar]
- 5.Recchia FM, Scott IU, Brown GC, Brown MM, Ho AC, Ip MS. Small-gauge pars plana vitrectomy: a report by the American Academy of Ophthalmology. Ophthalmology. 2010;117(9):1851–1857. doi: 10.1016/j.ophtha.2010.06.014. [DOI] [PubMed] [Google Scholar]
- 6.Schönfeld CL. 23- vs 20-gauge pars plana vitrectomy in combination with bimanual microincisional cataract surgery (b-MICS) for the treatment of macular hole and cataract as a one-step procedure. Eye (Lond) 2013;27(8):952–958. doi: 10.1038/eye.2013.118. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Krishnan R, Tossounis C, Fung Yang Y. 20-gauge and 23-gauge phacovitrectomy for idiopathic macular holes: comparison of complications and long-term outcomes. Eye (Lond) 2013;27(1):72–77. doi: 10.1038/eye.2012.227. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Can I, Takmaz T, Bayhan HA, Bostanci Ceran B. Aspheric microincision intraocular lens implantation with biaxial microincision cataract surgery: efficacy and reliability. J Cataract Refract Surg. 2010;36(11):1905–1911. doi: 10.1016/j.jcrs.2010.06.057. [DOI] [PubMed] [Google Scholar]
- 9.Podboraczyńska-Jodko K, Lubiński W. Bimanual microincision cataract surgery with implantation of an Akreos MI6O lens-one year follow-up. Klin Oczna. 2012;114(4):245–248. [PubMed] [Google Scholar]
- 10.Park DH, Shin JP, Kim SY. Combined 23-gauge microincisonal vitrectomy surgery and phacoemulsification with AcrySof toric intraocular lens implantation: a comparative study. Eye (Lond) 2011;25(10):1327–1332. doi: 10.1038/eye.2011.168. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Shinoda K, O'Hira A, Ishida S, Hoshide M, Ogawa LS, Ozawa Y, Nagasaki K, Inoue M, Katsura H. Posterior synechia of the iris after combined pars plana vitrectomy, phacoemulsification, and intraocular lens implantation. Jpn J Ophthalmol. 2001;45(3):276–280. doi: 10.1016/s0021-5155(01)00319-7. [DOI] [PubMed] [Google Scholar]
- 12.McDonnell PJ, Zarbin MA, Green WR. Posterior capsule opacification in pseudophakic eyes. Ophthalmology. 1983;90(12):1548–1553. doi: 10.1016/s0161-6420(83)34350-5. [DOI] [PubMed] [Google Scholar]
- 13.Yeo JH, Jakobiec FA, Pokorny K, Iwamoto T, Pisacano A. The ultrastructure of an IOL “cocoon membrane”. Ophthalmology. 1983;90(4):410–419. doi: 10.1016/s0161-6420(83)34546-2. [DOI] [PubMed] [Google Scholar]
- 14.Rudnisky CJ, Cheung J, Nanji JA. Intraocular lens capture in combined cataract extraction and pars plana vitrectomy: comparison of 1-piece and 3-piece acrylic intraocular lenses. J Cataract Refract Surg. 2010;36(10):1645–1649. doi: 10.1016/j.jcrs.2010.04.037. [DOI] [PubMed] [Google Scholar]
- 15.Honjo M, Ogura Y. Surgical results of pars plana vitrectomy combined with phacoemulsification and intraocular lens implantation for complications of proliferative diabetic retinopathy. Ophthalmic Surg Lasers. 1998;29(2):99–105. [PubMed] [Google Scholar]
- 16.Heiligenhaus A, Holtkamp A, Koch J, Schilling H, Bornfeld N, Losche CC, Steuhl KP. Combined phacoemulsification and pars plana vitrectomy: clear corneal versus scleral incisions: prospective randomized multicenter study. J Cataract Refract Surg. 2003;29(6):1106–1112. doi: 10.1016/s0886-3350(02)01811-4. [DOI] [PubMed] [Google Scholar]
- 17.Lahey JM, Francis RR, Kearney JJ. Combining phacoemulsification with pars plana vitrectomy in patients with proliferative diabetic retinopathy: a series of 223 cases. Ophthalmology. 2003;110(7):1335–1339. doi: 10.1016/S0161-6420(03)00454-8. [DOI] [PubMed] [Google Scholar]
- 18.Senn P, Schipper I, Perren B. Combined pars plana vitrectomy, phacoemulsification, and intraocular lens implantation in the capsular bag: a comparison to vitrectomy and subsequent cataract surgery as a two-step procedure. Ophthalmic Surg Lasers. 1995;26(5):420–428. [PubMed] [Google Scholar]
- 19.Amon M, Menapace R. In vivo documentation of cellular reactions on lens surfaces for assessing the biocompatibility of different intraocular implants. Eye (Lond) 1994;8(Pt 6):649–656. doi: 10.1038/eye.1994.161. [DOI] [PubMed] [Google Scholar]
- 20.Hanemoto T, Ideta H, Kawasaki T, Ideta R, Okuyama M. Suture fixation of an intraocular lens combined with pars plana vitrectomy and gas tamponade. J Cataract Refract Surg. 2003;29(12):2458–2460. doi: 10.1016/s0886-3350(03)00494-2. [DOI] [PubMed] [Google Scholar]
- 21.Iwase T, Sugiyama K. Investigation of the stability of one-piece acrylic intraocular lenses in cataract surgery and in combined vitrectomy surgery. Br J Ophthalmol. 2006;90(12):1519–1523. doi: 10.1136/bjo.2006.097766. [DOI] [PMC free article] [PubMed] [Google Scholar]