- Email: [email protected]
Pseudoexfoliation: High risk factors for zonule weakness and concurrent vitrectomy during phacoemulsification
ARTICLE
Pseudoexfoliation: High risk factors for zonule weakness and concurrent vitrectomy during phacoemulsification Bradford J. Shingleton, MD, Alicia C. Marvin, OD, Jeffrey S. Heier, MD, Mark W. O’Donoghue, OD, Anupam Laul, OD, Brian Wolff, OD, Anne Rowland, MD
PURPOSE: To assess the frequency and results of cataract surgery with concurrent vitrectomy due to zonule weakness in eyes with pseudoexfoliation (PXF). SETTING: Private practice, Boston, Massachusetts, USA. METHODS: This retrospective nonrandomized study assessed consecutive eyes with PXF having cataract surgery. High risk was defined as preoperative phacodonesis, iridodonesis, or lens subluxation (subgroup 1); preoperative anterior chamber depth or angle-depth asymmetry between eyes confirmed by slitlamp biomicroscopy or gonioscopy, respectively (subgroup 2); or complicated cataract extraction related to zonule weakness in the fellow eye (subgroup 3). Exclusion criteria were previous vitrectomy, scleral buckle, or trabeculectomy surgery and combined cataract– glaucoma surgery. A comparative analysis of outcome parameters was performed. RESULTS: Of the 1059 eyes evaluated, 38 had vitrectomy. Concurrent vitrectomy was required in 19 (15.6%) of 122 high-risk eyes and 19 (2.0%) of 937 non-high-risk eyes (P<.00001). The need for concurrent vitrectomy was greatest in subgroup 3 (72.7%) and subgroup 1 (42.9%). There was no statistically significant difference between the vitrectomy group and the no-vitrectomy group in the mean improvement in logMAR corrected distance visual acuity (CDVA) (P Z .38) or mean change in intraocular pressure (IOP) (mean decrease 2.6 mm Hg G 1.5 [SD] and 1.6 G 0.5 mm Hg, respectively) (P Z .47). CONCLUSION: Despite the need for vitrectomy and the attendant increased demands in postoperative care, both the vitrectomy group and no-vitrectomy group had improved logMAR CDVA and IOP. Financial Disclosure: No author has a financial or proprietary interest in any material or method mentioned. J Cataract Refract Surg 2010; 36:1261–1269 Q 2010 ASCRS and ESCRS
Pseudoexfoliation (PXF) is a systemic disorder with multiple ocular manifestations. Clinically, the disorder is characterized by the presence of white or gray flaky material on the anterior segment structures of the eye. The classic presentation is a bull’s-eye pattern of deposition on the anterior capsule of the lens. Pseudoexfoliation is the most identifiable cause of open-angle glaucoma.1,2 It can also be associated with a wide range of ocular issues, including cataract, zonule weakness, corneal decompensation, iris rigidity and poor iris dilation, and breakdown of the blood–aqueous barrier.2–4 In particular, zonule weakness can lead to increased complications during cataract surgery.2,5,6 Q 2010 ASCRS and ESCRS Published by Elsevier Inc.
Several studies7–16 report an increased rate of complications (eg, vitreous loss, zonule or capsule tears) during extracapsular cataract extraction (ECCE) in eyes with PXF. Studies of phacoemulsification report varying complication rates in eyes with PXF and eyes without PXF. Three studies17–19 found no significant differences between the 2 groups. However, 5 other studies20–24 found a statistically significant increase in complications in eyes with PXF compared with those in normal eyes. One study21 found that although the complications were more frequent in eyes with PXF than in eyes without, the long-term outcomes between the 2 groups were similar. Another study24 found equal intraoperative complications (in 0886-3350/$dsee front matter doi:10.1016/j.jcrs.2010.02.014
1261
1262
PHACOEMULSIFICATION AND VITRECTOMY IN PSEUDOEXFOLIATION
the absence of preoperative phacodonesis or subluxation) except for an increased rate of early postoperative intraocular pressure (IOP) spikes in eyes with PXF. These and other studies focus on features associated with PXF that might place the eye at a higher risk for complications during the course of cataract surgery. These features include anterior chamber depth (ACD) asymmetry, poor pupil dilation, zonule weakness, and the presence of phacodonesis. Ku¨chle et al.20 found decreased ACD in eyes with PXF was associated with an increase in complications. Ju¨nemann et al.A report a 13.4% risk for intraoperative complications in eyes with PXF that had an ACD less than 2.5 mm; eyes with an ACD greater than 2.5 mm had a 6.9% risk. According to Moreno et al.,25 the following signs are associated with zonule weakness in PXF: dense nuclear sclerosis, poor mydriasis, presence of glaucoma, atrophy of the pupillary frill, and uniform trabecular meshwork pigment. In a study of 1000 eyes that had ECCE, Guzek et al.26 found that a small pupil was a risk factor for vitreous loss and the presence of PXF was a risk factor for zonule weakness. In a preliminary report, Naumann et al.6,11 reported a 5-fold increase in vitreous loss with poor mydriasis and zonule fragility with PXF; later, they reported a 7-fold increase in vitreous loss. Alfaiate et al.27 found an increase in intraoperative complications associated with decreased pupil mydriasis in eyes with PXF. In a study of eyes with PXF, Pouliquen et al.15 found an increase in capsule rupture, zonule rupture, or both when the pupil was smaller than 6.0 mm; an increase in inflammation and complications postoperatively; and a decrease in final visual acuity. This study analyzed the visual and IOP outcomes in eyes with PXF that had phacoemulsification with concurrent vitrectomy because of zonule weakness. The study also compared the concurrent vitrectomy rate between eyes with PXF with and without preoperative risk factors for zonule weakness. PATIENTS AND METHODS This retrospective nonrandomized study assessed consecutive eyes with PXF that had cataract surgery by the same
Submitted: October 9, 2009. Final revision submitted: January 12, 2010. Accepted: February 5, 2010. From Center for Eye Research and Education, Boston, Massachusetts, USA. Corresponding author: Bradford J. Shingleton, MD, Ophthalmic Consultants of Boston, Center for Eye Research and Education, 50 Staniford Street, Suite 600, Boston, Massachusetts 02114, USA. E-mail: [email protected].
surgeon (B.J.S.) between 1987 and 2008. Results were compared with those in eyes in the same cohort that did not require vitrectomy. Eyes were excluded if they had previous vitrectomy, scleral buckle, or trabeculectomy surgery or combined cataract–glaucoma surgery. Outcome measures were logMAR corrected distance visual acuity (CDVA), IOP, glaucoma medication requirements, need for longterm topical antiinflammatory treatment, intraoperative and postoperative complications, and need for further surgery. Eyes at high risk for zonule weakness were identified by chart review and included those with preoperative phacodonesis, iridodonesis, or lens subluxation (subgroup 1); those with ACD or angle depth asymmetry within a given eye or between eyes (confirmed by slitlamp biomicroscopy or gonioscopy, respectively) that was not the result of a difference in axial length (AL) between eyes (subgroup 2); and those with complicated cataract extraction related to zonule weakness in the fellow eye (subgroup 3). The incidence of vitrectomy in high-risk eyes and non-high-risk eyes was compared for each decade during which the surgery was performed (1980s, 1990s, and 2000s) and for different age ranges (!65 years, 65 to 85 years, and O85 years). The incidences were compared using the hypothesis test of the equality of 2 proportions. Because pupil size was not documented preoperatively, it was not included as a risk factor. The presence, not the extent, of PXF material25 was documented in the study eyes. Although corneal endothelial changes can also occur in eyes with PXF, this parameter was not specifically documented or evaluated preoperatively. Phacoemulsification instrumentation and techniques evolved over the extended interval of the study; however, the basic principles of effective hydrodissection, segmental nuclear disassembly, and manual irrigation/aspiration of residual cortex were used in all cases. Comparisons were also made based on type of intraocular lens (IOL) implanted, planned versus unplanned vitrectomy, intraoperative timing of the vitrectomy, and type of vitrectomy. Planned vitrectomies were performed in eyes with extensive, preoperatively documented zonule weakness, such as cases of obvious lens subluxation. Unplanned vitrectomies were performed when zonule weakness was identified during phacoemulsification rather than preoperatively. SPSS software (SPSS, Inc.) was used to analyze means for statistical significance. A 2-tailed independent t test for equality of means was used to compare visual acuity, IOP, and glaucoma medication requirements in the groups. The results were confirmed by 1-way analysis of variance (ANOVA). Paired samples t tests were used to compare the preoperative and postoperative results within each group. The incidence rates of complications were analyzed using the hypothesis test of the equality of 2 proportions. Data are presented as means G SD. A P value of 0.05 or less was considered statistically significant. In the statistical analysis, histogram analysis showed that all data, except logMAR CDVA, were normally distributed. In both populations (concurrent vitrectomy and no concurrent vitrectomy), the logMAR CDVA was bimodally distributed, with 1 group distributed around 0.30 and the other around 1.25. The group centered at the logMAR vision of 1.25 represented outliers and numbered 218 preoperatively (204 in no-vitrectomy group, 14 in vitrectomy group) and 54 postoperatively (5 in vitrectomy group, 49 in novitrectomy group). When the majority of eyes (those with
J CATARACT REFRACT SURG - VOL 36, AUGUST 2010
1263
PHACOEMULSIFICATION AND VITRECTOMY IN PSEUDOEXFOLIATION
Table 1. Patient demographics. Parameter
Vitrectomy No Vitrectomy All Patients
Eyes (n) 38 Men (n) 6 Women (n) 32 Mean age (y) G SD 81.4 G 6.1 Mean follow-up 74.7 G 49.8 (mo) G SD
1021 293 728 79.3 G 6.9 21.3 G 31.6
1059 299 760 79.4 G 6.9 48.0 G 40.7
better logMAR CDVA) were analyzed separately, there was no change in any outcome measure (logMAR CDVA, IOP, number of glaucoma medications).
RESULTS Thirty-eight of 1059 eyes (3.6%) had concurrent cataract surgery and vitrectomy because of zonule weakness. Table 1 shows the demographics and follow-up data. On chart review, 122 eyes (11.5%) were determined to be at high risk for zonule weakness. Nineteen highrisk eyes (15.6%) and 19 (2.0%) of 937 non-high-risk eyes required concurrent vitrectomy; the difference was statistically significant (P!.00001). In the highrisk group, 21 eyes had preoperative phacodonesis, iridodonesis, or lens subluxation (subgroup 1), 90 had preoperative ACD or angle-degree asymmetry (subgroup 2), and 11 had complicated surgery in the fellow eye (subgroup 3). Nine eyes (42.9%) in subgroup 1, 2 eyes (2.2%) in subgroup 2, and 8 eyes (72.7%) in subgroup 3 required a vitrectomy. The difference in the vitrectomy rate between the 2 subgroups with the highest rate (1 and 2) was not statistically significant (PO.05). Combining the eyes in subgroup 1 and subgroup 3 yielded a vitrectomy rate of 53.1% (17/32); this incidence was significantly higher than in subgroup 2 (P!.01). A comparison of high-risk eyes showed a concurrent vitrectomy with cataract surgery was performed in
0 of 3 eyes in the 1980s (0%), 6 (2.6%) of 235 eyes in the 1990s, and 32 (3.9%) of 821 eyes in the 2000s. The difference between the decades was not statistically significant (PO.05). No patient younger than 65 years (n Z 20) had vitrectomy. Thirty-one (3.7%) of 841 eyes in patients between 65 years and 85 years and 7 (3.5%) of 198 eyes in patients older than 85 years old required vitrectomy. There were no statistically significant differences in the vitrectomy rate between the age groups (PO.05). Table 2 shows the preoperative-to-postoperative change in logMAR CDVA, IOP, and number of glaucoma medications in eyes requiring vitrectomy and eyes not requiring vitrectomy. The mean increase in CDVA was 0.40 G 0.01 logMAR in the vitrectomy group and 0.30 G 0.01 logMAR in the no-vitrectomy group; the mean decrease in IOP was 2.6 G 1.5 mm Hg and 1.6 G 0.5 mm Hg, respectively; and the mean decrease in the number of glaucoma medications was 0.10 G 0.20 and 0.00 G 0.01, respectively. There was no statistically significant difference between the 2 groups in logMAR CDVA, IOP, or glaucoma medications (P Z .38, P Z .47, and P Z .88, respectively). In most cases of poor preoperative visual acuity (R1.00 logMAR), the cause was a dense cataract (n Z 166). Other causes were fibrosis, atrophy, or macular scarring from age-related macular degeneration (AMD) (n Z 48); lens subluxation (n Z 1); a history of retinal detachments (n Z 1); and a history of vein occlusion (n Z 2). Postoperatively, logMAR CDVA improved in most of these eyes. In eyes with a postoperative CDVA worse than 1.00 logMAR, the causes were suprachoroidal hemorrhage (n Z 1), retinal detachment (n Z 1), history of preoperative retinal detachment (n Z 1), AMD (n Z 49), and past vein occlusion (n Z 2). Table 3 shows the preoperative-to-postoperative change in logMAR CDVA, IOP, and number of glaucoma medications by type of IOL; that is, anterior
Table 2. Comparison of preoperative and postoperative visual acuity, IOP, and glaucoma medications in the vitrectomy group and the no-vitrectomy group. Vitrectomy (n Z 38)
No Vitrectomy (n Z 1021)
Mean G SD Parameter CDVA (logMAR) IOP (mm Hg) Glaucoma medications (n)
Mean G SD
Preop
Postop
P Value
Preop
Postop
P Value
0.80 G 0.40 18.6 G 4.5 0.40 G 0.70
0.40 G 0.40 16.0 G 3.0 0.30 G 0.50
!.0001 .004 .83
0.60 G 0.04 16.8 G 3.9 0.30 G 0.08
0.30 G 0.03 15.2 G 4.4 0.30 G 0.07
!.0001 !.0001 .9
CDVA Z corrected distance visual acuity; IOP Z intraocular pressure
J CATARACT REFRACT SURG - VOL 36, AUGUST 2010
1264
PHACOEMULSIFICATION AND VITRECTOMY IN PSEUDOEXFOLIATION
Table 3. Comparison of preoperative and postoperative visual acuity, IOP, and glaucoma medications based on IOL type. Posterior Chamber IOL Anterior Chamber IOL (n Z 25)
Sutured in Sulcus (n Z 5)
Mean G SD Parameter CDVA (logMAR) IOP (mm Hg) Glaucoma medications (n)
Nonsutured in Capsule/Sulcus (n Z 5)
Mean G SD
Mean G SD
Preop
Postop
P Value
Preop
Postop
P Value
Preop
Postop
P Value
0.7 G 0.4 19.0 G 5.3 0.2 G 0.5
0.3 G 0.4 16 G 3.0 0.3 G 0.6
.001 .03 .38
0.7 G 0.4 17.0 G 3.1 0.4 G 0.5
0.5 G 0.2 13.6 G 1.8 0.2 G 0.4
.17 .07 .37
0.9 G 0.6 19.0 G 1.0 0.0 G 0.0
0.3 G 0.3 15.7 G 2.1 0.3 G 0.6
.09 .06 .04
CDVA Z corrected distance visual acuity; IOL Z intraocular lens; IOP Z intraocular pressure
chamber IOL (AC IOL), posterior chamber IOL (PC IOL) sutured in the sulcus, or nonsutured PC IOL fixated in the sulcus or capsular bag. The difference in logMAR CDVA and IOP between preoperatively and the last follow-up visit (mean 74.7 G 49.8 months) was statistically significant in the AC IOL group. The difference in glaucoma medications between preoperatively and the last follow-up was statistically significant in the nonsutured PC IOL group. There were no other differences between preoperatively and postoperatively in any parameter in any group. There was no statistically significant difference in logMAR CDVA, IOP, or number of glaucoma medications at the last follow-up visit between eyes with an AC IOL and eyes with a PC IOL. The independent t test showed a P value of 0.54 for logMAR CDVA, 0.12 for IOP, and 0.64 for glaucoma medications. The 1-way ANOVA showed P values of 0.69, 0.11, and 0.67, respectively. There was no statistically significant between-group difference in the preoperative-topostoperative change in any parameter (P Z .65, P Z .69, and P Z .44, respectively). Three of 38 eyes that had cataract surgery with concurrent vitrectomy were left aphakic. The mean preoperative CDVA was 1.2 G 0.6 logMAR; the
mean IOP, 17.5 G 2.1 mm Hg, and the mean number of glaucoma medications, 0.0 G 0.0. The postoperative means were 0.8 G 0.7 logMAR, 19.0 G 0.0 mm Hg, and 0.0 G 0.0, respectively. Neither the improvement in mean logMAR CDVA (P Z .39) nor the increase in mean IOP (P Z .5) was statistically significant. Table 4 shows the preoperative-to-postoperative changes in logMAR CDVA, IOP, and number of glaucoma medications categorized by eyes that had planned vitrectomy and eyes that had unplanned vitrectomy because of zonule weakness. There was no statistically significant difference between the 2 groups in any outcome measure at the last followup. The independent t test showed a P value of 0.75 for logMAR CDVA, 0.39 for IOP, and 0.37 for number for glaucoma medications. The 1-way ANOVA showed P values of 0.88, 0.59, and 0.38, respectively. There was no statistically significant between-group difference in the preoperative-to-postoperative change in any parameter (P Z .22, P Z .17, and P Z .36, respectively). Table 5 shows the preoperative-to-postoperative change in logMAR CDVA, IOP, and number of glaucoma medications categorized by eyes that had early conversion to vitrectomy (before initiation of
Table 4. Comparison of preoperative and postoperative visual acuity, IOP, and glaucoma medications between planned vitrectomy and unplanned vitrectomy. Planned Vitrectomy (n Z 12)
Unplanned Vitrectomy (n Z 26)
Mean G SD Parameter CDVA (logMAR) IOP (mm Hg) Glaucoma medications (n)
Mean G SD
Preop
Postop
P Value
Preop
Postop
P Value
0.9 G 0.4 19.1 G 5.5 0.7 G 1.0
0.38 G 0.4 15.5 G 3.5 0.5 G 0.5
.004 .03 .28
0.7 G 0.4 18.3 G 4.1 0.2 G 0.5
0.4 G 0.4 16.1 G 2.4 0.3 G 0.6
.002 .006 .61
CDVA Z corrected distance visual acuity; IOP Z intraocular pressure
J CATARACT REFRACT SURG - VOL 36, AUGUST 2010
1265
PHACOEMULSIFICATION AND VITRECTOMY IN PSEUDOEXFOLIATION
Table 5. Comparison of preoperative and postoperative visual acuity, IOP, and glaucoma medications based on time of conversion to vitrectomy. Early Conversion (n Z 27)
Late Conversion (n Z 11)
Mean G SD Parameter CDVA (logMAR) IOP (mm Hg) Glaucoma medications (n)
Mean G SD
Preop
Postop
P Value
Preop
Postop
P Value
0.7 G 0.4 19.0 G 5.0 0.4 G 0.8
0.4 G 0.4 16.5 G 3.0 0.3 G 0.6
!.0001 .01 .60
0.9 G 0.5 16.8 G 2.6 0.3 G 0.7
0.4 G 0.5 14.0 G 1.4 0.4 G 0.5
.02 .12 .73
CDVA Z corrected distance visual acuity; IOP Z intraocular pressure
phacoemulsification) and eyes that had late conversion (after initiation of phacoemulsification). There was no statistically significant difference between the 2 groups in logMAR CDVA or number of glaucoma medications at the last follow-up. The independent t test showed a P value of 0.74 for logMAR CDVA and 0.70 for glaucoma medications The 1-way ANOVA showed P values of 0.70 and 0.79, respectively. The independent t test showed a statistically significant greater reduction in IOP in eyes with late conversion (P Z .027); however, 1-way ANOVA test showed no statistically significant difference (P Z .07). There was no statistically significant between-group difference in the preoperative-to-postoperative change in any parameter (P Z .08, logMAR CDVA; P Z .08, IOP; and P Z .78, glaucoma medication). Table 6 shows the preoperative-to-postoperative change in logMAR CDVA, IOP, and number of glaucoma medications categorized by eyes that had anterior (limbal) vitrectomy and eyes that had posterior (pars plana) vitrectomy. There was no statistically significant difference between the 2 groups in any outcome measure at the last follow-up. The independent t test showed a P value of 0.9 for the logMAR CDVA, 0.45 for IOP, and 0.46 for glaucoma medication. The 1-way ANOVA showed P values of 0.68, 0.50, and 0.34, respectively.
An intraoperative complication occurred in 3 (7.9%) of the 38 eyes that had concurrent vitrectomy with cataract extraction; a suprachoroidal hemorrhage developed in 1 eye and iridodialysis in 2 eyes. All eyes with intraoperative complications were left aphakic at the time of the initial surgery. Two (66.6%) of the 3 eyes with intraoperative complications had an unplanned vitrectomy; 1 (33.3%) had a planned vitrectomy. One of the 3 eyes had early conversion to vitrectomy and 2 had late conversion. All 3 eyes had an anterior (limbal) vitrectomy. A postoperative complication occurred in 14 eyes (36.8%) with concurrent vitrectomy. The postoperative complications were cystoid macular edema (CME) (n Z 5); recurrent hyphema, vitreous hemorrhage, or pupillary block (n Z 2 eyes each); and persistent IOL displacement, corneal edema, or retinal detachment (n Z 1 eye each). Three eyes (21.4%) with a postoperative complication received a PC IOL, 8 (57.1%) received an AC IOL, and 3 (21.4%) were left aphakic. The rate of postoperative complications was statistically significantly higher in eyes with an AC IOL than in eyes with a PC IOL or eyes that were left aphakic (P!.05). Two eyes (14.2%) with a postoperative complication had planned vitrectomy and 12 (85.7%) had unplanned vitrectomy concurrent with cataract extraction; the difference between the 2
Table 6. Comparison of preoperative and postoperative visual acuity, IOP, and glaucoma medications based on vitrectomy approach. Anterior (Limbal) Vitrectomy (n Z 21)
Posterior (Pars Plana) Vitrectomy (n Z 17)
Mean G SD Parameter CDVA (logMAR) IOP (mm Hg) Glaucoma medications (n)
Mean G SD
Preop
Postop
P Value
Preop
Postop
P Value
0.7 G 0.5 18.0 G 3.7 0.3 G 0.8
0.4 G 0.4 16.3 G 2.3 0.4 G 0.6
.001 .17 .63
0.8 G 0.31 19.1 G 5.2 0.4 G 0.6
0.4 G 0.4 15.6 G 3.3 0.2 G 0.4
.001 .013 .19
CDVA Z corrected distance visual acuity; IOP Z intraocular pressure
J CATARACT REFRACT SURG - VOL 36, AUGUST 2010
1266
PHACOEMULSIFICATION AND VITRECTOMY IN PSEUDOEXFOLIATION
groups was statistically significant (P!.01). Five eyes (35.7%) with postoperative complications had late conversion to vitrectomy, and 9 (64.3%) had early conversion; the difference between the 2 groups was not statistically significant (PO.05). Five eyes (35.7%) with a postoperative complication had a posterior (pars plana) vitrectomy, and 9 (64.3%) had an anterior (limbal) vitrectomy; the difference between the 2 groups was not statistically significant (PO.05). Four (10.5%) of the 38 eyes that had phacoemulsification with concurrent vitrectomy required further laser or surgery or both after the initial cataract extraction. One eye required 2 anterior chamber washouts for recurrent hyphemas, a pars plana vitrectomy (PPV), and cryotherapy for recurrent vitreous hemorrhage as well as secondary IOL implantation (eye was originally aphakic). Two eyes required laser iris iridotomies (1 eye at 1 day and 1 eye at 1 month) for pupillary block; both eyes had an AC IOL. The fourth eye required a PPV, cryotherapy, and a gas bubble for retinal detachment. One eye required a neodymium:YAG (Nd:YAG) capsulotomy for posterior capsule opacification (PCO), but this was not included in our analysis of complications. Nineteen (50.0%) of 38 eyes that had phacoemulsification concurrent with vitrectomy required late-onset postoperative antiinflammatory topical therapy (topical prednisolone acetate 1% and/or topical nonsteroidal antiinflammatory drugs [NSAID]) for longer than 4 weeks. Eight eyes required a 6-week taper, 6 eyes required a 2-month taper, 3 eyes required a 3month course (2 for recurrent iritis, 1 for recurrent vitreous hemorrhage), 1 eye required a 4-month course for recurrent hyphema, and 1 eye required a 6-month course for corneal edema. Of these eyes, 13 (68.4%) received an AC IOL, 5 (26.3%) received a PC IOL, and 1 (5.3%) was left aphakic. Twelve eyes (63.2%) had an anterior limbal vitrectomy, and 7 (36.8%) had a PPV. Of the 1021 eyes (103 high risk) that did not have vitrectomy, 26 (2.5%; 12 high risk) had an intraoperative complication. Intraoperative complications were zonule weakness (n Z 22), need for a capsule tension ring (n Z 3), and a torn capsule (n Z 1). Of the 26 eyes, 5 had preoperative phacodonesis and 7 had preoperative ACD or angle-depth asymmetry. Of the eyes that did not have vitrectomy, 63 (6.2%) had a postoperative complication. Postoperative complications were IOL decentration, dislocation, or subluxation (n Z 18); iritis after 3 months (n Z 11); CME (n Z 9); increased AMD (n Z 7); retinal tear or detachment (n Z 6); retinal branch or central vein occlusion (n Z 3); and vitreous prolapse, recurrent corneal erosion, macular hole, need for laser for macular edema, need for panretinal photocoagulation for ocular ischemic syndrome and neovascular glaucoma,
incision leak, hyphema, need for an epiretinal membrane peel, and or corneal edema at 1 year (n Z 1 eye each). Of the 65 eyes (11 high risk) with postoperative complications that did not require concurrent vitrectomy, 3 had preoperative phacodonesis and 8 had preoperative ACD or angle-depth asymmetry. The difference in the incidence of postoperative complications was statistically significantly higher in eyes that required a vitrectomy (36.8%) than in eyes that did not (6.2%) (P!.01); however, the incidence of IOL decentration, dislocation, or subluxation was higher in the no-vitrectomy group. Thirty-six eyes (3.5%) that did not have vitrectomy required further laser treatment or surgery after the initial cataract extraction. The additional procedures included IOL reposition/exchange (n Z 11), retinal tear or detachment repair (n Z 6), argon laser trabeculoplasty or selective laser trabeculoplasty (n Z 6), injection or laser for AMD (n Z 2), anterior Nd:YAG capsulotomy for IOL decentration (n Z 4), filtering surgery (n Z 3), and laser for macular edema, panretinal photocoagulation ocular ischemic syndrome with neovascular glaucoma, peripheral iridotomy, or laser gonioplasty (n Z 1 eye each). The difference in incidence of additional surgery between eyes that required vitrectomy (13.2%) and eyes that did not (3.5%) was statistically significant (P!.01). An Nd:YAG laser capsulotomy for PCO was performed in 215 eyes (20.8%) in the no-vitrectomy group; these eyes with capsulotomy were not included in the analysis of postoperative complications. In the no-vitrectomy group, 34 eyes (3.3%) required a postoperative antiinflammatory regimen for more than 4 weeks or as a late-onset treatment. Eleven of these eyes had iritis after 3 months, 2 had a hyphema on the first postoperative day, 3 required a 6-week taper of antiinflammatory agents, 6 had an increase in AMD, 9 developed CME, and 3 had a retinal branch or central vein occlusion. The difference in the need for long-term antiinflammatory therapy between eyes that required a vitrectomy (50.0%) and those that did not (3.3%) was statistically significant (P!.01). DISCUSSION The presence of PXF increases the need for vitrectomy at the time of phacoemulsification surgery. In our series, 3.6% of eyes (38/1059) required concurrent vitrectomy due to zonule weakness. This rate is greater than reported in other populations without PXF.28,29 Of particular importance is the impact of preoperative risk factors on the rate of concurrent vitrectomy. In our study, we took into account several potential risk factors cited by previous authors,A,7,11,15,20,25–27 including preoperative phacodonesis, iridodonesis, or lens
J CATARACT REFRACT SURG - VOL 36, AUGUST 2010
PHACOEMULSIFICATION AND VITRECTOMY IN PSEUDOEXFOLIATION
subluxation; preoperative ACD asymmetry confirmed by slitlamp biomicroscopy or angle-depth asymmetry confirmed by gonioscopy that was not the result of a difference in AL between or within eyes; and a history of complicated surgery in the other eye due to zonule weakness. If these risk factors are taken into account, the incidence of vitrectomy in our study would increase to 15.6%, whereas the incidence would be 2.1% in eyes without these risk factors. A further analysis of potential significant risk factors showed that preoperative phacodonesis, iridodonesis, or lens subluxation and a history of complicated cataract surgery in the fellow eye significantly increased the risk for concurrent vitrectomy. The rate of vitrectomy in this group of patients was approximately 50%. Although ACD or angle-depth asymmetry did not prove to be a significant risk factor for intraoperative conversion to vitrectomy in our study, the presence of phacodonesis, iridodonesis, or subluxation or a history of complicated surgery in the fellow eye due to zonule weakness greatly increased the frequency of concurrent vitrectomy. It is important that anterior segment surgeons take these factors into account when considering cataract surgery in eyes with PXF. Even though concurrent vitrectomy is a significant event when combined with phacoemulsification, when given appropriate treatment, eyes that require concurrent vitrectomy and those that do not have similar improvement in logMAR CDVA and similar reduction in IOP and the need for glaucoma medication, as we found in our study. In addition, in our study, the logMAR CDVA, IOP, and glaucoma medication outcomes were not significantly different between eyes with an AC IOL and eyes with a PC IOL, whether sutured in the sulcus or nonsutured and implanted in the sulcus or capsular bag. Eyes that were left aphakic fared worse than those that received IOLs. This is probably because the surgeries in the aphakic group were significantly more complicated; an intraocular suprachoroidal hemorrhage occurred in 1 case and iridodialysis in 2 cases. Surgeon experience, as reflected in the number of vitrectomies required in each of the 3 decades studied in our review, did not affect the vitrectomy rate. In addition, patient age had no effect on the rate. Increasing patient age is often associated with increased cataract density and nucleus brunescence. Although cataract density per se was not evaluated as a risk factor in this study, older patients did not have a higher risk for concurrent vitrectomy, suggesting that cataract density did not play a major role in the risk for vitrectomy. Significantly more women than men required vitrectomy, and this increased incidence extends beyond the female preponderance of our population. The reason for the difference is unclear.
1267
Whether the vitrectomy was planned or unplanned did not have significant impact on postoperative logMAR CDVA, IOP, or need for glaucoma medication. However, a greater percentage of eyes that had unplanned vitrectomy had postoperative complications that required management. Unplanned surgical events are often more problematic to manage than planned procedures, which may be a reason for this result. Although the reduction in IOP was significantly greater in eyes that required a late conversion to vitrectomy, overall there was no significant difference in postoperative results between eyes that had early conversion and eyes that had late conversion to vitrectomy. A relatively similar number of eyes had an anterior (limbal) vitrectomy or a posterior (pars plana) vitrectomy. LogMAR CDVA, IOP, and glaucoma medication requirements improved in both groups, with no statistically significant differences between the 2 groups. There were no cases of vitreous incarceration in the anterior incisions. This suggests that wellperformed anterior vitrectomies can be equally successful as a pars plana approach. As one might expect, serious intraoperative complications were significantly greater in the eyes having vitrectomy than in the eyes that did not have vitrectomy. An intraoperative suprachoroidal hemorrhage or significant iridodialysis occurred in approximately 8% of the 38 eyes in the vitrectomy group. Intraocular lens implantation was not possible in these eyes, and postoperative management was more intensive. The complications occurred because of the extensive manipulation required to express nuclear fragments before the vitrectomy. Despite the significant improvement in logMAR CDVA, IOP reduction, and stability of glaucoma medication requirements postoperatively in eyes that required concurrent vitrectomy with phacoemulsification, the incidence of postoperative complications was greater than in eyes that did not require concurrent vitrectomy. Cystoid macular edema was the most common complication (13.2%). These patients were managed with topical steroids and NSAIDs; with treatment, the CME resolved and logMAR CDVA improved in 3 (60.0%) of the 5 eyes. Of the remaining 2 eyes, 1 had no resolution of the CME and the other was lost to follow-up. Other complications, including recurrent hyphema, vitreous hemorrhage, pupillary block, and corneal edema, did not occur in the novitrectomy group. These complications are not unexpected given the increased manipulation and duration of surgery associated with phacoemulsification and concurrent vitrectomy. Cystoid macular edema and retinal detachment occur more frequently in vitrectomized eyes than in eyes with an intact posterior
J CATARACT REFRACT SURG - VOL 36, AUGUST 2010
1268
PHACOEMULSIFICATION AND VITRECTOMY IN PSEUDOEXFOLIATION
capsule. Pupillary block is more common in eyes without a posterior capsule, corneal edema is likely related to greater posterior segment manipulation, and bleeding issues are likely related to greater posterior segment manipulation.30,31 Of note, early or late postoperative PC IOL decentration, dislocation, or subluxation occurred in 18 (1.7%) of 1021 eyes in the no-vitrectomy group, and 11 required IOL repositioning or exchange. This rate of IOL displacement is not unexpected given that 22 eyes (2.1%) in the novitrectomy group had confirmed zonule weakness at the time of surgery and it is well documented that late dislocation can occur in eyes with PXF postoperatively, presumably as a result of progressive zonulysis.4 Although 1 eye in the vitrectomy group had PC IOL displacement, no eye in that group required later IOL surgery. This might be expected because in the vitrectomy group, IOL positioning was secured at the time of initial surgery by anterior chamber placement or sulcus suturing. Although the rate of postoperative complications was higher in eyes with an AC IOL than in eyes with a PC IOL, it is unlikely that the complications per se were related to IOL type because there were no specific problems related to IOL placement or positioning. Despite the higher postoperative complication rate in eyes with an AC IOL, there were no significant differences in the final logMAR CDVA, IOP, or number of glaucoma medications between the AC IOL group and PC IOL group. Five eyes (13.2%) in the vitrectomy group required subsequent surgery to correct problems such as recurrent hyphema, vitreous hemorrhage, secondary implantation for surgical aphakia, laser iridectomy for pupillary block, and posterior segment surgery for retinal detachment. This rate was significantly higher than in eyes not requiring vitrectomy. Despite the significantly greater trauma associated with vitrectomy at the time of phacoemulsification, chronic or recurrent inflammation was not a major problem in eyes having a vitrectomy. Fifty percent of the eyes required topical antiinflammatory therapy (topical steroid agents or NSAIDs) beyond the typical 4-week period required in noncomplicated eyes. However, all except 3 eyes, which had recurrent hyphema/ vitreous hemorrhage, prolonged corneal edema, or chronic CME, had improved vision after treatment. In summary, all surgeons strive to avoid the complication of vitrectomy. In eyes with PXF, the incidence of concurrent vitrectomy is greater than in eyes without the disorder. Thus, surgeons should pay particular attention to the role of risk factors in the rate of vitrectomy. Eyes with PXF associated with phacodonesis, iridodonesis, lens subluxation, or complicated surgery in the fellow eye have about a 50% chance of requiring concurrent vitrectomy with cataract surgery.
Despite the higher vitrectomy rates and increased intraoperative and postoperative complications associated with vitrectomy, patients having concurrent phacoemulsification and vitrectomy due to zonule weakness can often achieve satisfactory postoperative vision, IOP, and glaucoma medication results when they receive appropriate care and follow-up. REFERENCES 1. Ritch R. Exfoliationdthe most common identifiable cause of open-angle glaucoma. J Glaucoma 1994; 3:176–178 2. Drolsum L, Ringvold A, Nicolaissen B. Cataract and glaucoma surgery in pseudoexfoliation syndrome: a review. Acta Ophthalmologica Scand 2007; 85:810–821. Available at: http://www3. interscience.wiley.com/cgi-bin/fulltext/118515723/PDFSTART. Accessed April 12, 2010 3. Conway RM, Schlo¨tzer-Schrehardt U, Ku¨chle M, Naumann GOH. Pseudoexfoliation syndrome: pathological manifestations of relevance to intraocular surgery. Clin Exp Ophthalmol 2004; 32:199–210 4. Schlo¨tzer-Schrehardt U, Naumann GOH. A histopathologic study of zonular instability in pseudoexfoliation syndrome. Am J Ophthalmol 1994; 118:730–743 5. Shingleton BJ, Crandall AS, Ahmed K II. Pseudoexfoliation and the cataract surgeon: preoperative, intraoperative, and postoperative issues related to intraocular pressure, cataract, and intraocular lenses. J Cataract Refract Surg 2009; 35:1101–1120 6. Naumann GOH. Exfoliation as a risk factor for vitreous loss in extracapsular cataract surgery (preliminary report); the ‘‘ErlangerAugenbluten Group.’’. Acta Ophthalmol Suppl 1988; 184:129– 131. Available at: http://www3.interscience.wiley.com/cgi-bin/ fulltext/122411524/PDFSTART. Accessed April 12, 2010 7. Davis D, Brubaker J, Espander L, Stringham J, Crandall A, Werner L, Mamalis N. Late in-the-bag spontaneous intraocular lens dislocation; evaluation of 86 consecutive cases. Ophthalmology 2009; 116:664–670 8. Skuta GL, Parrish RK II, Hodapp E, Forster RK, Rockwood EJ. Zonular dialysis during extracapsular cataract extraction in pseudoexfoliation syndrome. Arch Ophthalmol 1987; 105:632– 634 9. Lumme P, Laatikainen L. Exfoliation syndrome and cataract extraction. Am J Ophthalmol 1993; 116:51–55 10. Avramides S, Traianidis P, Sakkias G. Cataract surgery and lens implantation in eyes with exfoliation syndrome. J Cataract Refract Surg 1997; 23:583–587 11. Naumann GOH, Ku¨chle M, Scho¨nherr U. Pseudoexfoliationssyndrom als Risikofaktor fu¨r Glasko¨rperverlust bei der extrakakpsula¨ren Kataraktextraktion; Erlanger Augenbla¨tter-Gruppe [Pseudoexfoliation syndrome as a risk factor for vitreous loss in extracapsular cataract surgery; the Erlanger Ophthalmology Group]. Fortschr Ophthalmol 1989; 86:543–545 12. Kirkpatrick JNP, Harrad RA. Complicated extracapsular cataract surgery in pseudoexfoliation syndrome: a case report. Br J Ophthalmol 1992; 76:692–693. Available at: http://www.ncbi. nlm.nih.gov/pmc/articles/PMC504379/pdf/brjopthal00059-0052. pdf. Accessed April 12, 2010 13. Ku¨chle M, Scho¨nherr U, Dieckmann U. Risikofaktoren fu¨r Kapselruptur und Glasko¨rperverlust bei extrakapsula¨rer Kataraktextraktion; die Erlangen Augenbla¨tter-Gruppe [Risk factors for capsular rupture and vitreous loss in extracapsular cataract extraction; the Erlangen Ophthalmology Group]. Fortschr Ophthalmologica 1989; 86:417–421 14. Zetterstro¨m C, Olivestedt G, Lundvall A. Exfoliation syndrome and extracapsular cataract extraction with implantation of
J CATARACT REFRACT SURG - VOL 36, AUGUST 2010
PHACOEMULSIFICATION AND VITRECTOMY IN PSEUDOEXFOLIATION
15.
16.
17.
18.
19.
20.
21.
22.
23.
24. 25.
posterior chamber lens. Acta Ophthalmol (Copenh) 1992; 70:85–90. Available at: http://www3.interscience.wiley.com/ cgi-bin/fulltext/122405906/PDFSTART. Accessed April 12, 2010 Pouliquen P, Robinet A, Colin J, Cochener B, Charpentier D. Syndrome exfoliatif et chirurgie de la cataracte [Exfoliative syndrome and cataract surgery]. J Fr Ophtalmol 1992; 15:171–176 Goder GJ. Our experiences in planned extracapsular cataract extraction in the exfoliation syndrome. Acta Ophthalmol Suppl (Copenh) 1988; 184:126–128. Available at: http://www3. interscience.wiley.com/cgi-bin/fulltext/122411503/PDFSTART. Accessed April 12, 2010 Shastri L, Vasvada A. Phacoemulsification in Indian eyes with pseudoexfoliation syndrome. J Cataract Refract Surg 2001; 27:1629–1637 Hyams M, Mathalone N, Herskovitz M, Hod Y, Israeli D, Geyer O. Intraoperative complications in phacoemulsification in eyes with and without pseudoexfoliation. J Cataract Refract Surg 2005; 31:1002–1005 Dosso AA, Bonvin ER, Leuenberger PM. Exfoliation syndrome and phacoemulsification. J Cataract Refract Surg 1997; 23:122–125 Ku¨chle M, Viestenz A, Martus P, Ha¨ndel A, Ju¨nemann A, Naumann GOH. Anterior chamber depth and complications during cataract surgery in eyes with pseudoexfoliation syndrome. Am J Ophthalmol 2000; 129:281–285 Drolsum L, Haaskjold E, Sandvig K. Phacoemulsification in eyes with pseudoexfoliation. J Cataract Refract Surg 1998; 24:787– 792 Scorolli L, Scorolli L, Campos EC, Bassein L, Meduri RA. Pseudoexfoliation syndrome: a cohort study on intraoperative complications in cataract surgery. Ophthalmologica 1998; 212:278–280 Hayashi H, Hayashi K, Nakao F, Hayashi F. Anterior capsule contraction and intraocular lens dislocation in eyes with pseudoexfoliation syndrome. Br J Ophthalmol 1998; 82:1429–1432. Available at: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1722461/pdf/ v082p01429.pdf. Accessed April 12, 2010 Akinci A, Batman C, Zilelioglu O. Phacoemulsification in pseudoexfoliation syndrome. Ophthalmologica 2008; 222:112–116 Moreno J, Duch S, Lajara J. Pseudoexfoliation syndrome: clinical factors related to capsular rupture in cataract surgery. Acta Ophthalmol (Copenh) 1993; 71:181–184. Available at: http://www3. interscience.wiley.com/cgi-bin/fulltext/122435497/PDFSTART. Accessed April 12, 2010
1269
26. Guzek JP, Holm M, Cotter JB, Cameron JA, Rademaker WJ, Wissinger DH, Tonjum AM, Sleeper LA. Risk factors for intraocular complications in 1000 extracapsular cataract cases. Ophthalmology 1987; 94:461–466 27. Alfaiate M, Leite E, Mira J, Cunha-Vaz JG. Prevalence and surgical complications of pseudoexfoliation syndrome in Portuguese patients with senile cataract. J Cataract Refract Surg 1996; 22:972–976 28. Pot MC, Stilma JS. Laag complicatierisico bij cataractoperaties uitgevoerd door artsen in opleiding tot oogarts. [Low complication rate with cataract operations carried out by registrars in ophthalmology]. Ned Tijschr Geneeskd 2008; 152:563–568 29. Zaidi FH, Corbett MC, Burton BJL, Bloom PA. Raising the benchmark for the 21st centurydthe 1000 cataract operations audit and survey: outcomes, consultant-supervised training and sourcing NHS choice. Br J Ophthalmol 2007; 91:731–736. Table 2 correction available at: http://bjo.bmj.com/content/ suppl/2007/05/30/bjo.2006.104216.DC1/916731webonlyfig. pdf. Accessed April 12, 2010 30. Chan FM, Mathur R, Ku JJK, Chen C, Chan S-P, Yong VSH. Au Eong K-G. Short term outcomes in eyes with posterior capsule rupture during cataract surgery. J Cataract Refract Surg 2003; 29:531–541 31. Jakobsson G, Montan P, Zetterberg M, Stenevi U, Behndig A, Lundstro¨m M. Capsule complication during cataract surgery: retinal detachment after cataract surgery with capsule complication; Swedish Capsule Rupture Study Group report 4. J Cataract Refract Surg 2009; 35:1699–1705
OTHER CITED MATERIAL A. Ju¨nemann A, Martus P, Handel A, Naumann GOH. ‘‘Ocular dimensions in pseudoexfoliation syndrome,’’ presented at the Joint European Meeting in Ophthalmology and Vision, Montpellier, France, October 1997. Abstract in Ophthalmic Res 1997; 29(suppl): abstract 340
J CATARACT REFRACT SURG - VOL 36, AUGUST 2010
First author: Bradford J. Shingleton, MD Center for Eye Research and Education, Boston, Massachusetts, USA
Pseudoexfoliation: High risk factors for zonule weakness and concurrent vitrectomy during phacoemulsification Bradford J. Shingleton, MD, Alicia C. Marvin, OD, Jeffrey S. Heier, MD, Mark W. O’Donoghue, OD, Anupam Laul, OD, Brian Wolff, OD, Anne Rowland, MD
PURPOSE: To assess the frequency and results of cataract surgery with concurrent vitrectomy due to zonule weakness in eyes with pseudoexfoliation (PXF). SETTING: Private practice, Boston, Massachusetts, USA. METHODS: This retrospective nonrandomized study assessed consecutive eyes with PXF having cataract surgery. High risk was defined as preoperative phacodonesis, iridodonesis, or lens subluxation (subgroup 1); preoperative anterior chamber depth or angle-depth asymmetry between eyes confirmed by slitlamp biomicroscopy or gonioscopy, respectively (subgroup 2); or complicated cataract extraction related to zonule weakness in the fellow eye (subgroup 3). Exclusion criteria were previous vitrectomy, scleral buckle, or trabeculectomy surgery and combined cataract– glaucoma surgery. A comparative analysis of outcome parameters was performed. RESULTS: Of the 1059 eyes evaluated, 38 had vitrectomy. Concurrent vitrectomy was required in 19 (15.6%) of 122 high-risk eyes and 19 (2.0%) of 937 non-high-risk eyes (P<.00001). The need for concurrent vitrectomy was greatest in subgroup 3 (72.7%) and subgroup 1 (42.9%). There was no statistically significant difference between the vitrectomy group and the no-vitrectomy group in the mean improvement in logMAR corrected distance visual acuity (CDVA) (P Z .38) or mean change in intraocular pressure (IOP) (mean decrease 2.6 mm Hg G 1.5 [SD] and 1.6 G 0.5 mm Hg, respectively) (P Z .47). CONCLUSION: Despite the need for vitrectomy and the attendant increased demands in postoperative care, both the vitrectomy group and no-vitrectomy group had improved logMAR CDVA and IOP. Financial Disclosure: No author has a financial or proprietary interest in any material or method mentioned. J Cataract Refract Surg 2010; 36:1261–1269 Q 2010 ASCRS and ESCRS
Pseudoexfoliation (PXF) is a systemic disorder with multiple ocular manifestations. Clinically, the disorder is characterized by the presence of white or gray flaky material on the anterior segment structures of the eye. The classic presentation is a bull’s-eye pattern of deposition on the anterior capsule of the lens. Pseudoexfoliation is the most identifiable cause of open-angle glaucoma.1,2 It can also be associated with a wide range of ocular issues, including cataract, zonule weakness, corneal decompensation, iris rigidity and poor iris dilation, and breakdown of the blood–aqueous barrier.2–4 In particular, zonule weakness can lead to increased complications during cataract surgery.2,5,6 Q 2010 ASCRS and ESCRS Published by Elsevier Inc.
Several studies7–16 report an increased rate of complications (eg, vitreous loss, zonule or capsule tears) during extracapsular cataract extraction (ECCE) in eyes with PXF. Studies of phacoemulsification report varying complication rates in eyes with PXF and eyes without PXF. Three studies17–19 found no significant differences between the 2 groups. However, 5 other studies20–24 found a statistically significant increase in complications in eyes with PXF compared with those in normal eyes. One study21 found that although the complications were more frequent in eyes with PXF than in eyes without, the long-term outcomes between the 2 groups were similar. Another study24 found equal intraoperative complications (in 0886-3350/$dsee front matter doi:10.1016/j.jcrs.2010.02.014
1261
1262
PHACOEMULSIFICATION AND VITRECTOMY IN PSEUDOEXFOLIATION
the absence of preoperative phacodonesis or subluxation) except for an increased rate of early postoperative intraocular pressure (IOP) spikes in eyes with PXF. These and other studies focus on features associated with PXF that might place the eye at a higher risk for complications during the course of cataract surgery. These features include anterior chamber depth (ACD) asymmetry, poor pupil dilation, zonule weakness, and the presence of phacodonesis. Ku¨chle et al.20 found decreased ACD in eyes with PXF was associated with an increase in complications. Ju¨nemann et al.A report a 13.4% risk for intraoperative complications in eyes with PXF that had an ACD less than 2.5 mm; eyes with an ACD greater than 2.5 mm had a 6.9% risk. According to Moreno et al.,25 the following signs are associated with zonule weakness in PXF: dense nuclear sclerosis, poor mydriasis, presence of glaucoma, atrophy of the pupillary frill, and uniform trabecular meshwork pigment. In a study of 1000 eyes that had ECCE, Guzek et al.26 found that a small pupil was a risk factor for vitreous loss and the presence of PXF was a risk factor for zonule weakness. In a preliminary report, Naumann et al.6,11 reported a 5-fold increase in vitreous loss with poor mydriasis and zonule fragility with PXF; later, they reported a 7-fold increase in vitreous loss. Alfaiate et al.27 found an increase in intraoperative complications associated with decreased pupil mydriasis in eyes with PXF. In a study of eyes with PXF, Pouliquen et al.15 found an increase in capsule rupture, zonule rupture, or both when the pupil was smaller than 6.0 mm; an increase in inflammation and complications postoperatively; and a decrease in final visual acuity. This study analyzed the visual and IOP outcomes in eyes with PXF that had phacoemulsification with concurrent vitrectomy because of zonule weakness. The study also compared the concurrent vitrectomy rate between eyes with PXF with and without preoperative risk factors for zonule weakness. PATIENTS AND METHODS This retrospective nonrandomized study assessed consecutive eyes with PXF that had cataract surgery by the same
Submitted: October 9, 2009. Final revision submitted: January 12, 2010. Accepted: February 5, 2010. From Center for Eye Research and Education, Boston, Massachusetts, USA. Corresponding author: Bradford J. Shingleton, MD, Ophthalmic Consultants of Boston, Center for Eye Research and Education, 50 Staniford Street, Suite 600, Boston, Massachusetts 02114, USA. E-mail: [email protected].
surgeon (B.J.S.) between 1987 and 2008. Results were compared with those in eyes in the same cohort that did not require vitrectomy. Eyes were excluded if they had previous vitrectomy, scleral buckle, or trabeculectomy surgery or combined cataract–glaucoma surgery. Outcome measures were logMAR corrected distance visual acuity (CDVA), IOP, glaucoma medication requirements, need for longterm topical antiinflammatory treatment, intraoperative and postoperative complications, and need for further surgery. Eyes at high risk for zonule weakness were identified by chart review and included those with preoperative phacodonesis, iridodonesis, or lens subluxation (subgroup 1); those with ACD or angle depth asymmetry within a given eye or between eyes (confirmed by slitlamp biomicroscopy or gonioscopy, respectively) that was not the result of a difference in axial length (AL) between eyes (subgroup 2); and those with complicated cataract extraction related to zonule weakness in the fellow eye (subgroup 3). The incidence of vitrectomy in high-risk eyes and non-high-risk eyes was compared for each decade during which the surgery was performed (1980s, 1990s, and 2000s) and for different age ranges (!65 years, 65 to 85 years, and O85 years). The incidences were compared using the hypothesis test of the equality of 2 proportions. Because pupil size was not documented preoperatively, it was not included as a risk factor. The presence, not the extent, of PXF material25 was documented in the study eyes. Although corneal endothelial changes can also occur in eyes with PXF, this parameter was not specifically documented or evaluated preoperatively. Phacoemulsification instrumentation and techniques evolved over the extended interval of the study; however, the basic principles of effective hydrodissection, segmental nuclear disassembly, and manual irrigation/aspiration of residual cortex were used in all cases. Comparisons were also made based on type of intraocular lens (IOL) implanted, planned versus unplanned vitrectomy, intraoperative timing of the vitrectomy, and type of vitrectomy. Planned vitrectomies were performed in eyes with extensive, preoperatively documented zonule weakness, such as cases of obvious lens subluxation. Unplanned vitrectomies were performed when zonule weakness was identified during phacoemulsification rather than preoperatively. SPSS software (SPSS, Inc.) was used to analyze means for statistical significance. A 2-tailed independent t test for equality of means was used to compare visual acuity, IOP, and glaucoma medication requirements in the groups. The results were confirmed by 1-way analysis of variance (ANOVA). Paired samples t tests were used to compare the preoperative and postoperative results within each group. The incidence rates of complications were analyzed using the hypothesis test of the equality of 2 proportions. Data are presented as means G SD. A P value of 0.05 or less was considered statistically significant. In the statistical analysis, histogram analysis showed that all data, except logMAR CDVA, were normally distributed. In both populations (concurrent vitrectomy and no concurrent vitrectomy), the logMAR CDVA was bimodally distributed, with 1 group distributed around 0.30 and the other around 1.25. The group centered at the logMAR vision of 1.25 represented outliers and numbered 218 preoperatively (204 in no-vitrectomy group, 14 in vitrectomy group) and 54 postoperatively (5 in vitrectomy group, 49 in novitrectomy group). When the majority of eyes (those with
J CATARACT REFRACT SURG - VOL 36, AUGUST 2010
1263
PHACOEMULSIFICATION AND VITRECTOMY IN PSEUDOEXFOLIATION
Table 1. Patient demographics. Parameter
Vitrectomy No Vitrectomy All Patients
Eyes (n) 38 Men (n) 6 Women (n) 32 Mean age (y) G SD 81.4 G 6.1 Mean follow-up 74.7 G 49.8 (mo) G SD
1021 293 728 79.3 G 6.9 21.3 G 31.6
1059 299 760 79.4 G 6.9 48.0 G 40.7
better logMAR CDVA) were analyzed separately, there was no change in any outcome measure (logMAR CDVA, IOP, number of glaucoma medications).
RESULTS Thirty-eight of 1059 eyes (3.6%) had concurrent cataract surgery and vitrectomy because of zonule weakness. Table 1 shows the demographics and follow-up data. On chart review, 122 eyes (11.5%) were determined to be at high risk for zonule weakness. Nineteen highrisk eyes (15.6%) and 19 (2.0%) of 937 non-high-risk eyes required concurrent vitrectomy; the difference was statistically significant (P!.00001). In the highrisk group, 21 eyes had preoperative phacodonesis, iridodonesis, or lens subluxation (subgroup 1), 90 had preoperative ACD or angle-degree asymmetry (subgroup 2), and 11 had complicated surgery in the fellow eye (subgroup 3). Nine eyes (42.9%) in subgroup 1, 2 eyes (2.2%) in subgroup 2, and 8 eyes (72.7%) in subgroup 3 required a vitrectomy. The difference in the vitrectomy rate between the 2 subgroups with the highest rate (1 and 2) was not statistically significant (PO.05). Combining the eyes in subgroup 1 and subgroup 3 yielded a vitrectomy rate of 53.1% (17/32); this incidence was significantly higher than in subgroup 2 (P!.01). A comparison of high-risk eyes showed a concurrent vitrectomy with cataract surgery was performed in
0 of 3 eyes in the 1980s (0%), 6 (2.6%) of 235 eyes in the 1990s, and 32 (3.9%) of 821 eyes in the 2000s. The difference between the decades was not statistically significant (PO.05). No patient younger than 65 years (n Z 20) had vitrectomy. Thirty-one (3.7%) of 841 eyes in patients between 65 years and 85 years and 7 (3.5%) of 198 eyes in patients older than 85 years old required vitrectomy. There were no statistically significant differences in the vitrectomy rate between the age groups (PO.05). Table 2 shows the preoperative-to-postoperative change in logMAR CDVA, IOP, and number of glaucoma medications in eyes requiring vitrectomy and eyes not requiring vitrectomy. The mean increase in CDVA was 0.40 G 0.01 logMAR in the vitrectomy group and 0.30 G 0.01 logMAR in the no-vitrectomy group; the mean decrease in IOP was 2.6 G 1.5 mm Hg and 1.6 G 0.5 mm Hg, respectively; and the mean decrease in the number of glaucoma medications was 0.10 G 0.20 and 0.00 G 0.01, respectively. There was no statistically significant difference between the 2 groups in logMAR CDVA, IOP, or glaucoma medications (P Z .38, P Z .47, and P Z .88, respectively). In most cases of poor preoperative visual acuity (R1.00 logMAR), the cause was a dense cataract (n Z 166). Other causes were fibrosis, atrophy, or macular scarring from age-related macular degeneration (AMD) (n Z 48); lens subluxation (n Z 1); a history of retinal detachments (n Z 1); and a history of vein occlusion (n Z 2). Postoperatively, logMAR CDVA improved in most of these eyes. In eyes with a postoperative CDVA worse than 1.00 logMAR, the causes were suprachoroidal hemorrhage (n Z 1), retinal detachment (n Z 1), history of preoperative retinal detachment (n Z 1), AMD (n Z 49), and past vein occlusion (n Z 2). Table 3 shows the preoperative-to-postoperative change in logMAR CDVA, IOP, and number of glaucoma medications by type of IOL; that is, anterior
Table 2. Comparison of preoperative and postoperative visual acuity, IOP, and glaucoma medications in the vitrectomy group and the no-vitrectomy group. Vitrectomy (n Z 38)
No Vitrectomy (n Z 1021)
Mean G SD Parameter CDVA (logMAR) IOP (mm Hg) Glaucoma medications (n)
Mean G SD
Preop
Postop
P Value
Preop
Postop
P Value
0.80 G 0.40 18.6 G 4.5 0.40 G 0.70
0.40 G 0.40 16.0 G 3.0 0.30 G 0.50
!.0001 .004 .83
0.60 G 0.04 16.8 G 3.9 0.30 G 0.08
0.30 G 0.03 15.2 G 4.4 0.30 G 0.07
!.0001 !.0001 .9
CDVA Z corrected distance visual acuity; IOP Z intraocular pressure
J CATARACT REFRACT SURG - VOL 36, AUGUST 2010
1264
PHACOEMULSIFICATION AND VITRECTOMY IN PSEUDOEXFOLIATION
Table 3. Comparison of preoperative and postoperative visual acuity, IOP, and glaucoma medications based on IOL type. Posterior Chamber IOL Anterior Chamber IOL (n Z 25)
Sutured in Sulcus (n Z 5)
Mean G SD Parameter CDVA (logMAR) IOP (mm Hg) Glaucoma medications (n)
Nonsutured in Capsule/Sulcus (n Z 5)
Mean G SD
Mean G SD
Preop
Postop
P Value
Preop
Postop
P Value
Preop
Postop
P Value
0.7 G 0.4 19.0 G 5.3 0.2 G 0.5
0.3 G 0.4 16 G 3.0 0.3 G 0.6
.001 .03 .38
0.7 G 0.4 17.0 G 3.1 0.4 G 0.5
0.5 G 0.2 13.6 G 1.8 0.2 G 0.4
.17 .07 .37
0.9 G 0.6 19.0 G 1.0 0.0 G 0.0
0.3 G 0.3 15.7 G 2.1 0.3 G 0.6
.09 .06 .04
CDVA Z corrected distance visual acuity; IOL Z intraocular lens; IOP Z intraocular pressure
chamber IOL (AC IOL), posterior chamber IOL (PC IOL) sutured in the sulcus, or nonsutured PC IOL fixated in the sulcus or capsular bag. The difference in logMAR CDVA and IOP between preoperatively and the last follow-up visit (mean 74.7 G 49.8 months) was statistically significant in the AC IOL group. The difference in glaucoma medications between preoperatively and the last follow-up was statistically significant in the nonsutured PC IOL group. There were no other differences between preoperatively and postoperatively in any parameter in any group. There was no statistically significant difference in logMAR CDVA, IOP, or number of glaucoma medications at the last follow-up visit between eyes with an AC IOL and eyes with a PC IOL. The independent t test showed a P value of 0.54 for logMAR CDVA, 0.12 for IOP, and 0.64 for glaucoma medications. The 1-way ANOVA showed P values of 0.69, 0.11, and 0.67, respectively. There was no statistically significant between-group difference in the preoperative-topostoperative change in any parameter (P Z .65, P Z .69, and P Z .44, respectively). Three of 38 eyes that had cataract surgery with concurrent vitrectomy were left aphakic. The mean preoperative CDVA was 1.2 G 0.6 logMAR; the
mean IOP, 17.5 G 2.1 mm Hg, and the mean number of glaucoma medications, 0.0 G 0.0. The postoperative means were 0.8 G 0.7 logMAR, 19.0 G 0.0 mm Hg, and 0.0 G 0.0, respectively. Neither the improvement in mean logMAR CDVA (P Z .39) nor the increase in mean IOP (P Z .5) was statistically significant. Table 4 shows the preoperative-to-postoperative changes in logMAR CDVA, IOP, and number of glaucoma medications categorized by eyes that had planned vitrectomy and eyes that had unplanned vitrectomy because of zonule weakness. There was no statistically significant difference between the 2 groups in any outcome measure at the last followup. The independent t test showed a P value of 0.75 for logMAR CDVA, 0.39 for IOP, and 0.37 for number for glaucoma medications. The 1-way ANOVA showed P values of 0.88, 0.59, and 0.38, respectively. There was no statistically significant between-group difference in the preoperative-to-postoperative change in any parameter (P Z .22, P Z .17, and P Z .36, respectively). Table 5 shows the preoperative-to-postoperative change in logMAR CDVA, IOP, and number of glaucoma medications categorized by eyes that had early conversion to vitrectomy (before initiation of
Table 4. Comparison of preoperative and postoperative visual acuity, IOP, and glaucoma medications between planned vitrectomy and unplanned vitrectomy. Planned Vitrectomy (n Z 12)
Unplanned Vitrectomy (n Z 26)
Mean G SD Parameter CDVA (logMAR) IOP (mm Hg) Glaucoma medications (n)
Mean G SD
Preop
Postop
P Value
Preop
Postop
P Value
0.9 G 0.4 19.1 G 5.5 0.7 G 1.0
0.38 G 0.4 15.5 G 3.5 0.5 G 0.5
.004 .03 .28
0.7 G 0.4 18.3 G 4.1 0.2 G 0.5
0.4 G 0.4 16.1 G 2.4 0.3 G 0.6
.002 .006 .61
CDVA Z corrected distance visual acuity; IOP Z intraocular pressure
J CATARACT REFRACT SURG - VOL 36, AUGUST 2010
1265
PHACOEMULSIFICATION AND VITRECTOMY IN PSEUDOEXFOLIATION
Table 5. Comparison of preoperative and postoperative visual acuity, IOP, and glaucoma medications based on time of conversion to vitrectomy. Early Conversion (n Z 27)
Late Conversion (n Z 11)
Mean G SD Parameter CDVA (logMAR) IOP (mm Hg) Glaucoma medications (n)
Mean G SD
Preop
Postop
P Value
Preop
Postop
P Value
0.7 G 0.4 19.0 G 5.0 0.4 G 0.8
0.4 G 0.4 16.5 G 3.0 0.3 G 0.6
!.0001 .01 .60
0.9 G 0.5 16.8 G 2.6 0.3 G 0.7
0.4 G 0.5 14.0 G 1.4 0.4 G 0.5
.02 .12 .73
CDVA Z corrected distance visual acuity; IOP Z intraocular pressure
phacoemulsification) and eyes that had late conversion (after initiation of phacoemulsification). There was no statistically significant difference between the 2 groups in logMAR CDVA or number of glaucoma medications at the last follow-up. The independent t test showed a P value of 0.74 for logMAR CDVA and 0.70 for glaucoma medications The 1-way ANOVA showed P values of 0.70 and 0.79, respectively. The independent t test showed a statistically significant greater reduction in IOP in eyes with late conversion (P Z .027); however, 1-way ANOVA test showed no statistically significant difference (P Z .07). There was no statistically significant between-group difference in the preoperative-to-postoperative change in any parameter (P Z .08, logMAR CDVA; P Z .08, IOP; and P Z .78, glaucoma medication). Table 6 shows the preoperative-to-postoperative change in logMAR CDVA, IOP, and number of glaucoma medications categorized by eyes that had anterior (limbal) vitrectomy and eyes that had posterior (pars plana) vitrectomy. There was no statistically significant difference between the 2 groups in any outcome measure at the last follow-up. The independent t test showed a P value of 0.9 for the logMAR CDVA, 0.45 for IOP, and 0.46 for glaucoma medication. The 1-way ANOVA showed P values of 0.68, 0.50, and 0.34, respectively.
An intraoperative complication occurred in 3 (7.9%) of the 38 eyes that had concurrent vitrectomy with cataract extraction; a suprachoroidal hemorrhage developed in 1 eye and iridodialysis in 2 eyes. All eyes with intraoperative complications were left aphakic at the time of the initial surgery. Two (66.6%) of the 3 eyes with intraoperative complications had an unplanned vitrectomy; 1 (33.3%) had a planned vitrectomy. One of the 3 eyes had early conversion to vitrectomy and 2 had late conversion. All 3 eyes had an anterior (limbal) vitrectomy. A postoperative complication occurred in 14 eyes (36.8%) with concurrent vitrectomy. The postoperative complications were cystoid macular edema (CME) (n Z 5); recurrent hyphema, vitreous hemorrhage, or pupillary block (n Z 2 eyes each); and persistent IOL displacement, corneal edema, or retinal detachment (n Z 1 eye each). Three eyes (21.4%) with a postoperative complication received a PC IOL, 8 (57.1%) received an AC IOL, and 3 (21.4%) were left aphakic. The rate of postoperative complications was statistically significantly higher in eyes with an AC IOL than in eyes with a PC IOL or eyes that were left aphakic (P!.05). Two eyes (14.2%) with a postoperative complication had planned vitrectomy and 12 (85.7%) had unplanned vitrectomy concurrent with cataract extraction; the difference between the 2
Table 6. Comparison of preoperative and postoperative visual acuity, IOP, and glaucoma medications based on vitrectomy approach. Anterior (Limbal) Vitrectomy (n Z 21)
Posterior (Pars Plana) Vitrectomy (n Z 17)
Mean G SD Parameter CDVA (logMAR) IOP (mm Hg) Glaucoma medications (n)
Mean G SD
Preop
Postop
P Value
Preop
Postop
P Value
0.7 G 0.5 18.0 G 3.7 0.3 G 0.8
0.4 G 0.4 16.3 G 2.3 0.4 G 0.6
.001 .17 .63
0.8 G 0.31 19.1 G 5.2 0.4 G 0.6
0.4 G 0.4 15.6 G 3.3 0.2 G 0.4
.001 .013 .19
CDVA Z corrected distance visual acuity; IOP Z intraocular pressure
J CATARACT REFRACT SURG - VOL 36, AUGUST 2010
1266
PHACOEMULSIFICATION AND VITRECTOMY IN PSEUDOEXFOLIATION
groups was statistically significant (P!.01). Five eyes (35.7%) with postoperative complications had late conversion to vitrectomy, and 9 (64.3%) had early conversion; the difference between the 2 groups was not statistically significant (PO.05). Five eyes (35.7%) with a postoperative complication had a posterior (pars plana) vitrectomy, and 9 (64.3%) had an anterior (limbal) vitrectomy; the difference between the 2 groups was not statistically significant (PO.05). Four (10.5%) of the 38 eyes that had phacoemulsification with concurrent vitrectomy required further laser or surgery or both after the initial cataract extraction. One eye required 2 anterior chamber washouts for recurrent hyphemas, a pars plana vitrectomy (PPV), and cryotherapy for recurrent vitreous hemorrhage as well as secondary IOL implantation (eye was originally aphakic). Two eyes required laser iris iridotomies (1 eye at 1 day and 1 eye at 1 month) for pupillary block; both eyes had an AC IOL. The fourth eye required a PPV, cryotherapy, and a gas bubble for retinal detachment. One eye required a neodymium:YAG (Nd:YAG) capsulotomy for posterior capsule opacification (PCO), but this was not included in our analysis of complications. Nineteen (50.0%) of 38 eyes that had phacoemulsification concurrent with vitrectomy required late-onset postoperative antiinflammatory topical therapy (topical prednisolone acetate 1% and/or topical nonsteroidal antiinflammatory drugs [NSAID]) for longer than 4 weeks. Eight eyes required a 6-week taper, 6 eyes required a 2-month taper, 3 eyes required a 3month course (2 for recurrent iritis, 1 for recurrent vitreous hemorrhage), 1 eye required a 4-month course for recurrent hyphema, and 1 eye required a 6-month course for corneal edema. Of these eyes, 13 (68.4%) received an AC IOL, 5 (26.3%) received a PC IOL, and 1 (5.3%) was left aphakic. Twelve eyes (63.2%) had an anterior limbal vitrectomy, and 7 (36.8%) had a PPV. Of the 1021 eyes (103 high risk) that did not have vitrectomy, 26 (2.5%; 12 high risk) had an intraoperative complication. Intraoperative complications were zonule weakness (n Z 22), need for a capsule tension ring (n Z 3), and a torn capsule (n Z 1). Of the 26 eyes, 5 had preoperative phacodonesis and 7 had preoperative ACD or angle-depth asymmetry. Of the eyes that did not have vitrectomy, 63 (6.2%) had a postoperative complication. Postoperative complications were IOL decentration, dislocation, or subluxation (n Z 18); iritis after 3 months (n Z 11); CME (n Z 9); increased AMD (n Z 7); retinal tear or detachment (n Z 6); retinal branch or central vein occlusion (n Z 3); and vitreous prolapse, recurrent corneal erosion, macular hole, need for laser for macular edema, need for panretinal photocoagulation for ocular ischemic syndrome and neovascular glaucoma,
incision leak, hyphema, need for an epiretinal membrane peel, and or corneal edema at 1 year (n Z 1 eye each). Of the 65 eyes (11 high risk) with postoperative complications that did not require concurrent vitrectomy, 3 had preoperative phacodonesis and 8 had preoperative ACD or angle-depth asymmetry. The difference in the incidence of postoperative complications was statistically significantly higher in eyes that required a vitrectomy (36.8%) than in eyes that did not (6.2%) (P!.01); however, the incidence of IOL decentration, dislocation, or subluxation was higher in the no-vitrectomy group. Thirty-six eyes (3.5%) that did not have vitrectomy required further laser treatment or surgery after the initial cataract extraction. The additional procedures included IOL reposition/exchange (n Z 11), retinal tear or detachment repair (n Z 6), argon laser trabeculoplasty or selective laser trabeculoplasty (n Z 6), injection or laser for AMD (n Z 2), anterior Nd:YAG capsulotomy for IOL decentration (n Z 4), filtering surgery (n Z 3), and laser for macular edema, panretinal photocoagulation ocular ischemic syndrome with neovascular glaucoma, peripheral iridotomy, or laser gonioplasty (n Z 1 eye each). The difference in incidence of additional surgery between eyes that required vitrectomy (13.2%) and eyes that did not (3.5%) was statistically significant (P!.01). An Nd:YAG laser capsulotomy for PCO was performed in 215 eyes (20.8%) in the no-vitrectomy group; these eyes with capsulotomy were not included in the analysis of postoperative complications. In the no-vitrectomy group, 34 eyes (3.3%) required a postoperative antiinflammatory regimen for more than 4 weeks or as a late-onset treatment. Eleven of these eyes had iritis after 3 months, 2 had a hyphema on the first postoperative day, 3 required a 6-week taper of antiinflammatory agents, 6 had an increase in AMD, 9 developed CME, and 3 had a retinal branch or central vein occlusion. The difference in the need for long-term antiinflammatory therapy between eyes that required a vitrectomy (50.0%) and those that did not (3.3%) was statistically significant (P!.01). DISCUSSION The presence of PXF increases the need for vitrectomy at the time of phacoemulsification surgery. In our series, 3.6% of eyes (38/1059) required concurrent vitrectomy due to zonule weakness. This rate is greater than reported in other populations without PXF.28,29 Of particular importance is the impact of preoperative risk factors on the rate of concurrent vitrectomy. In our study, we took into account several potential risk factors cited by previous authors,A,7,11,15,20,25–27 including preoperative phacodonesis, iridodonesis, or lens
J CATARACT REFRACT SURG - VOL 36, AUGUST 2010
PHACOEMULSIFICATION AND VITRECTOMY IN PSEUDOEXFOLIATION
subluxation; preoperative ACD asymmetry confirmed by slitlamp biomicroscopy or angle-depth asymmetry confirmed by gonioscopy that was not the result of a difference in AL between or within eyes; and a history of complicated surgery in the other eye due to zonule weakness. If these risk factors are taken into account, the incidence of vitrectomy in our study would increase to 15.6%, whereas the incidence would be 2.1% in eyes without these risk factors. A further analysis of potential significant risk factors showed that preoperative phacodonesis, iridodonesis, or lens subluxation and a history of complicated cataract surgery in the fellow eye significantly increased the risk for concurrent vitrectomy. The rate of vitrectomy in this group of patients was approximately 50%. Although ACD or angle-depth asymmetry did not prove to be a significant risk factor for intraoperative conversion to vitrectomy in our study, the presence of phacodonesis, iridodonesis, or subluxation or a history of complicated surgery in the fellow eye due to zonule weakness greatly increased the frequency of concurrent vitrectomy. It is important that anterior segment surgeons take these factors into account when considering cataract surgery in eyes with PXF. Even though concurrent vitrectomy is a significant event when combined with phacoemulsification, when given appropriate treatment, eyes that require concurrent vitrectomy and those that do not have similar improvement in logMAR CDVA and similar reduction in IOP and the need for glaucoma medication, as we found in our study. In addition, in our study, the logMAR CDVA, IOP, and glaucoma medication outcomes were not significantly different between eyes with an AC IOL and eyes with a PC IOL, whether sutured in the sulcus or nonsutured and implanted in the sulcus or capsular bag. Eyes that were left aphakic fared worse than those that received IOLs. This is probably because the surgeries in the aphakic group were significantly more complicated; an intraocular suprachoroidal hemorrhage occurred in 1 case and iridodialysis in 2 cases. Surgeon experience, as reflected in the number of vitrectomies required in each of the 3 decades studied in our review, did not affect the vitrectomy rate. In addition, patient age had no effect on the rate. Increasing patient age is often associated with increased cataract density and nucleus brunescence. Although cataract density per se was not evaluated as a risk factor in this study, older patients did not have a higher risk for concurrent vitrectomy, suggesting that cataract density did not play a major role in the risk for vitrectomy. Significantly more women than men required vitrectomy, and this increased incidence extends beyond the female preponderance of our population. The reason for the difference is unclear.
1267
Whether the vitrectomy was planned or unplanned did not have significant impact on postoperative logMAR CDVA, IOP, or need for glaucoma medication. However, a greater percentage of eyes that had unplanned vitrectomy had postoperative complications that required management. Unplanned surgical events are often more problematic to manage than planned procedures, which may be a reason for this result. Although the reduction in IOP was significantly greater in eyes that required a late conversion to vitrectomy, overall there was no significant difference in postoperative results between eyes that had early conversion and eyes that had late conversion to vitrectomy. A relatively similar number of eyes had an anterior (limbal) vitrectomy or a posterior (pars plana) vitrectomy. LogMAR CDVA, IOP, and glaucoma medication requirements improved in both groups, with no statistically significant differences between the 2 groups. There were no cases of vitreous incarceration in the anterior incisions. This suggests that wellperformed anterior vitrectomies can be equally successful as a pars plana approach. As one might expect, serious intraoperative complications were significantly greater in the eyes having vitrectomy than in the eyes that did not have vitrectomy. An intraoperative suprachoroidal hemorrhage or significant iridodialysis occurred in approximately 8% of the 38 eyes in the vitrectomy group. Intraocular lens implantation was not possible in these eyes, and postoperative management was more intensive. The complications occurred because of the extensive manipulation required to express nuclear fragments before the vitrectomy. Despite the significant improvement in logMAR CDVA, IOP reduction, and stability of glaucoma medication requirements postoperatively in eyes that required concurrent vitrectomy with phacoemulsification, the incidence of postoperative complications was greater than in eyes that did not require concurrent vitrectomy. Cystoid macular edema was the most common complication (13.2%). These patients were managed with topical steroids and NSAIDs; with treatment, the CME resolved and logMAR CDVA improved in 3 (60.0%) of the 5 eyes. Of the remaining 2 eyes, 1 had no resolution of the CME and the other was lost to follow-up. Other complications, including recurrent hyphema, vitreous hemorrhage, pupillary block, and corneal edema, did not occur in the novitrectomy group. These complications are not unexpected given the increased manipulation and duration of surgery associated with phacoemulsification and concurrent vitrectomy. Cystoid macular edema and retinal detachment occur more frequently in vitrectomized eyes than in eyes with an intact posterior
J CATARACT REFRACT SURG - VOL 36, AUGUST 2010
1268
PHACOEMULSIFICATION AND VITRECTOMY IN PSEUDOEXFOLIATION
capsule. Pupillary block is more common in eyes without a posterior capsule, corneal edema is likely related to greater posterior segment manipulation, and bleeding issues are likely related to greater posterior segment manipulation.30,31 Of note, early or late postoperative PC IOL decentration, dislocation, or subluxation occurred in 18 (1.7%) of 1021 eyes in the no-vitrectomy group, and 11 required IOL repositioning or exchange. This rate of IOL displacement is not unexpected given that 22 eyes (2.1%) in the novitrectomy group had confirmed zonule weakness at the time of surgery and it is well documented that late dislocation can occur in eyes with PXF postoperatively, presumably as a result of progressive zonulysis.4 Although 1 eye in the vitrectomy group had PC IOL displacement, no eye in that group required later IOL surgery. This might be expected because in the vitrectomy group, IOL positioning was secured at the time of initial surgery by anterior chamber placement or sulcus suturing. Although the rate of postoperative complications was higher in eyes with an AC IOL than in eyes with a PC IOL, it is unlikely that the complications per se were related to IOL type because there were no specific problems related to IOL placement or positioning. Despite the higher postoperative complication rate in eyes with an AC IOL, there were no significant differences in the final logMAR CDVA, IOP, or number of glaucoma medications between the AC IOL group and PC IOL group. Five eyes (13.2%) in the vitrectomy group required subsequent surgery to correct problems such as recurrent hyphema, vitreous hemorrhage, secondary implantation for surgical aphakia, laser iridectomy for pupillary block, and posterior segment surgery for retinal detachment. This rate was significantly higher than in eyes not requiring vitrectomy. Despite the significantly greater trauma associated with vitrectomy at the time of phacoemulsification, chronic or recurrent inflammation was not a major problem in eyes having a vitrectomy. Fifty percent of the eyes required topical antiinflammatory therapy (topical steroid agents or NSAIDs) beyond the typical 4-week period required in noncomplicated eyes. However, all except 3 eyes, which had recurrent hyphema/ vitreous hemorrhage, prolonged corneal edema, or chronic CME, had improved vision after treatment. In summary, all surgeons strive to avoid the complication of vitrectomy. In eyes with PXF, the incidence of concurrent vitrectomy is greater than in eyes without the disorder. Thus, surgeons should pay particular attention to the role of risk factors in the rate of vitrectomy. Eyes with PXF associated with phacodonesis, iridodonesis, lens subluxation, or complicated surgery in the fellow eye have about a 50% chance of requiring concurrent vitrectomy with cataract surgery.
Despite the higher vitrectomy rates and increased intraoperative and postoperative complications associated with vitrectomy, patients having concurrent phacoemulsification and vitrectomy due to zonule weakness can often achieve satisfactory postoperative vision, IOP, and glaucoma medication results when they receive appropriate care and follow-up. REFERENCES 1. Ritch R. Exfoliationdthe most common identifiable cause of open-angle glaucoma. J Glaucoma 1994; 3:176–178 2. Drolsum L, Ringvold A, Nicolaissen B. Cataract and glaucoma surgery in pseudoexfoliation syndrome: a review. Acta Ophthalmologica Scand 2007; 85:810–821. Available at: http://www3. interscience.wiley.com/cgi-bin/fulltext/118515723/PDFSTART. Accessed April 12, 2010 3. Conway RM, Schlo¨tzer-Schrehardt U, Ku¨chle M, Naumann GOH. Pseudoexfoliation syndrome: pathological manifestations of relevance to intraocular surgery. Clin Exp Ophthalmol 2004; 32:199–210 4. Schlo¨tzer-Schrehardt U, Naumann GOH. A histopathologic study of zonular instability in pseudoexfoliation syndrome. Am J Ophthalmol 1994; 118:730–743 5. Shingleton BJ, Crandall AS, Ahmed K II. Pseudoexfoliation and the cataract surgeon: preoperative, intraoperative, and postoperative issues related to intraocular pressure, cataract, and intraocular lenses. J Cataract Refract Surg 2009; 35:1101–1120 6. Naumann GOH. Exfoliation as a risk factor for vitreous loss in extracapsular cataract surgery (preliminary report); the ‘‘ErlangerAugenbluten Group.’’. Acta Ophthalmol Suppl 1988; 184:129– 131. Available at: http://www3.interscience.wiley.com/cgi-bin/ fulltext/122411524/PDFSTART. Accessed April 12, 2010 7. Davis D, Brubaker J, Espander L, Stringham J, Crandall A, Werner L, Mamalis N. Late in-the-bag spontaneous intraocular lens dislocation; evaluation of 86 consecutive cases. Ophthalmology 2009; 116:664–670 8. Skuta GL, Parrish RK II, Hodapp E, Forster RK, Rockwood EJ. Zonular dialysis during extracapsular cataract extraction in pseudoexfoliation syndrome. Arch Ophthalmol 1987; 105:632– 634 9. Lumme P, Laatikainen L. Exfoliation syndrome and cataract extraction. Am J Ophthalmol 1993; 116:51–55 10. Avramides S, Traianidis P, Sakkias G. Cataract surgery and lens implantation in eyes with exfoliation syndrome. J Cataract Refract Surg 1997; 23:583–587 11. Naumann GOH, Ku¨chle M, Scho¨nherr U. Pseudoexfoliationssyndrom als Risikofaktor fu¨r Glasko¨rperverlust bei der extrakakpsula¨ren Kataraktextraktion; Erlanger Augenbla¨tter-Gruppe [Pseudoexfoliation syndrome as a risk factor for vitreous loss in extracapsular cataract surgery; the Erlanger Ophthalmology Group]. Fortschr Ophthalmol 1989; 86:543–545 12. Kirkpatrick JNP, Harrad RA. Complicated extracapsular cataract surgery in pseudoexfoliation syndrome: a case report. Br J Ophthalmol 1992; 76:692–693. Available at: http://www.ncbi. nlm.nih.gov/pmc/articles/PMC504379/pdf/brjopthal00059-0052. pdf. Accessed April 12, 2010 13. Ku¨chle M, Scho¨nherr U, Dieckmann U. Risikofaktoren fu¨r Kapselruptur und Glasko¨rperverlust bei extrakapsula¨rer Kataraktextraktion; die Erlangen Augenbla¨tter-Gruppe [Risk factors for capsular rupture and vitreous loss in extracapsular cataract extraction; the Erlangen Ophthalmology Group]. Fortschr Ophthalmologica 1989; 86:417–421 14. Zetterstro¨m C, Olivestedt G, Lundvall A. Exfoliation syndrome and extracapsular cataract extraction with implantation of
J CATARACT REFRACT SURG - VOL 36, AUGUST 2010
PHACOEMULSIFICATION AND VITRECTOMY IN PSEUDOEXFOLIATION
15.
16.
17.
18.
19.
20.
21.
22.
23.
24. 25.
posterior chamber lens. Acta Ophthalmol (Copenh) 1992; 70:85–90. Available at: http://www3.interscience.wiley.com/ cgi-bin/fulltext/122405906/PDFSTART. Accessed April 12, 2010 Pouliquen P, Robinet A, Colin J, Cochener B, Charpentier D. Syndrome exfoliatif et chirurgie de la cataracte [Exfoliative syndrome and cataract surgery]. J Fr Ophtalmol 1992; 15:171–176 Goder GJ. Our experiences in planned extracapsular cataract extraction in the exfoliation syndrome. Acta Ophthalmol Suppl (Copenh) 1988; 184:126–128. Available at: http://www3. interscience.wiley.com/cgi-bin/fulltext/122411503/PDFSTART. Accessed April 12, 2010 Shastri L, Vasvada A. Phacoemulsification in Indian eyes with pseudoexfoliation syndrome. J Cataract Refract Surg 2001; 27:1629–1637 Hyams M, Mathalone N, Herskovitz M, Hod Y, Israeli D, Geyer O. Intraoperative complications in phacoemulsification in eyes with and without pseudoexfoliation. J Cataract Refract Surg 2005; 31:1002–1005 Dosso AA, Bonvin ER, Leuenberger PM. Exfoliation syndrome and phacoemulsification. J Cataract Refract Surg 1997; 23:122–125 Ku¨chle M, Viestenz A, Martus P, Ha¨ndel A, Ju¨nemann A, Naumann GOH. Anterior chamber depth and complications during cataract surgery in eyes with pseudoexfoliation syndrome. Am J Ophthalmol 2000; 129:281–285 Drolsum L, Haaskjold E, Sandvig K. Phacoemulsification in eyes with pseudoexfoliation. J Cataract Refract Surg 1998; 24:787– 792 Scorolli L, Scorolli L, Campos EC, Bassein L, Meduri RA. Pseudoexfoliation syndrome: a cohort study on intraoperative complications in cataract surgery. Ophthalmologica 1998; 212:278–280 Hayashi H, Hayashi K, Nakao F, Hayashi F. Anterior capsule contraction and intraocular lens dislocation in eyes with pseudoexfoliation syndrome. Br J Ophthalmol 1998; 82:1429–1432. Available at: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1722461/pdf/ v082p01429.pdf. Accessed April 12, 2010 Akinci A, Batman C, Zilelioglu O. Phacoemulsification in pseudoexfoliation syndrome. Ophthalmologica 2008; 222:112–116 Moreno J, Duch S, Lajara J. Pseudoexfoliation syndrome: clinical factors related to capsular rupture in cataract surgery. Acta Ophthalmol (Copenh) 1993; 71:181–184. Available at: http://www3. interscience.wiley.com/cgi-bin/fulltext/122435497/PDFSTART. Accessed April 12, 2010
1269
26. Guzek JP, Holm M, Cotter JB, Cameron JA, Rademaker WJ, Wissinger DH, Tonjum AM, Sleeper LA. Risk factors for intraocular complications in 1000 extracapsular cataract cases. Ophthalmology 1987; 94:461–466 27. Alfaiate M, Leite E, Mira J, Cunha-Vaz JG. Prevalence and surgical complications of pseudoexfoliation syndrome in Portuguese patients with senile cataract. J Cataract Refract Surg 1996; 22:972–976 28. Pot MC, Stilma JS. Laag complicatierisico bij cataractoperaties uitgevoerd door artsen in opleiding tot oogarts. [Low complication rate with cataract operations carried out by registrars in ophthalmology]. Ned Tijschr Geneeskd 2008; 152:563–568 29. Zaidi FH, Corbett MC, Burton BJL, Bloom PA. Raising the benchmark for the 21st centurydthe 1000 cataract operations audit and survey: outcomes, consultant-supervised training and sourcing NHS choice. Br J Ophthalmol 2007; 91:731–736. Table 2 correction available at: http://bjo.bmj.com/content/ suppl/2007/05/30/bjo.2006.104216.DC1/916731webonlyfig. pdf. Accessed April 12, 2010 30. Chan FM, Mathur R, Ku JJK, Chen C, Chan S-P, Yong VSH. Au Eong K-G. Short term outcomes in eyes with posterior capsule rupture during cataract surgery. J Cataract Refract Surg 2003; 29:531–541 31. Jakobsson G, Montan P, Zetterberg M, Stenevi U, Behndig A, Lundstro¨m M. Capsule complication during cataract surgery: retinal detachment after cataract surgery with capsule complication; Swedish Capsule Rupture Study Group report 4. J Cataract Refract Surg 2009; 35:1699–1705
OTHER CITED MATERIAL A. Ju¨nemann A, Martus P, Handel A, Naumann GOH. ‘‘Ocular dimensions in pseudoexfoliation syndrome,’’ presented at the Joint European Meeting in Ophthalmology and Vision, Montpellier, France, October 1997. Abstract in Ophthalmic Res 1997; 29(suppl): abstract 340
J CATARACT REFRACT SURG - VOL 36, AUGUST 2010
First author: Bradford J. Shingleton, MD Center for Eye Research and Education, Boston, Massachusetts, USA