Risk factors for posterior capsular pearling after uncomplicated extracapsular cataract extraction and plano-convex posterior chamber lens implantation

a r t

-c

I

I es

Risk factors for posterior capsular pearling after uncomplicated extracapsular cataract extraction and plano-convex posterior chamber lens implantation Samir A. Jamal, M.D., Leon D. Solomon, F.R.C.S.(C)

ABSTRACT We retrospectively reviewed 542 consecutive cases of extracapsular cataract extraction with implantation of a plano-convex laser ridge posterior chamber intraocular lens. We were interested in the incidence and probability of the development of postcapsular epithelial pearling that required laser capsulotomy and the risk factors associated with it. After four years of follow-up, 21 % of all cases had developed pearling. Over the same period, the probability of pearling, as determined by the Kaplan-Meier curves, was 29% with a 95% confidence interval between 22% and 36%. After one year of follow-up, the probability of pearling was 5%; after two years, 15%; after three years, 24%. Younger age (50 years or less), larger optic, and smaller IOL were identified as significant risk factors when each of these factors was analyzed separately. When all risk factors were studied together, and after applying confounding statistics, only younger age was identified as a significant risk factor (P = .003). Key Words: capsular fibrosis, capsular pearling, extracapsular cataract extraction, optic size, posterior capsule, posterior capsule opacification, posterior chamber intraocular lens, YAG laser caps ulotomy

Posterior capsule opacification is a common problem after extracapsular cataract extraction (ECCE) with or without posterior chamber intraocular lens (IOL) implantation. Reported incidence varies from 8 % to 50%.1-4 A YAG laser capsulotomy has been used for almost a decade to treat posterior capsule opacification; however, this procedure is not without problems. Known complications include transient5,6 or persistent raised intraocular pressure/,8 uveitis,6 retinal detachment,9,10 cystoid macular edema,5 IOL injury,5 corneal endothelial changes and decompensation, 6 hyphema, 5 staphylococcal endophthalmitis, II chronic indolent infection by P_acne,12-14 macular holes,t5 ciliochoroidal effu-

sion,16 iris neovascularization and neovascular glaucoma in diabetics, 17 and central retinal artery occlusion. 6,18 In addition to these potential complications, the cost of a capsulotomy has a significant economic effect on the total cost of cataract services. The ophthalmic community is interested in reducing the frequency of peading of the posterior capsule and thus reducing the need for YAG laser capsulotomy and its potential complications. This can only be achieved by identifying the risk factors and then avoiding them if possible. We retrospectively reviewed the medical records of all private cases of one author (L.S.) that had several types of ECCE with posterior chamber IOL implantation. The

From Jewish General Hospital, McGill University, Quebec, Canada. Presented in part at the 14th Annual McGill Day. Montreal General Hospital. Montreal. Quebec. Canada. June 1991. Reprint requests to Dr. Leon D. Solomon. 5775 Cote des Neiges. Suite 107. Montreal. Quebec. Canada H3S 2S9. J CATARACT REFRACT SURG-VOL 19, MAY 1993

333

endocapsular technique was used in the following ways: standard manual ECCE; soft cataract aspiration; phacoemulsification. Our objectives were (1) to determine the incidence of late posterior capsule epithelial pearling requiring laser capsulotomy in patients who received a plano-convex, angled haptic, laser ridge IOL; (2) to identify risk factors for the development of posterior capsule epithelial pearling after uncomplicated ECCE and posterior chamber IOL, including age, sex, surgeon, surgical technique, and IOL features.

SUBJECfS AND METHODS

evidence of central pearling, (2) a two or more line decrease on the Snellen visual acuity chart that was accounted for by the presence of pearling, or (3) both. We used statistical techniques to handle censored data. 19 To compare the incidence of pearling between groups the log-rank test was used. Kaplan-Meier estimates with their associated confidence intervals were computed to estimate the probability of pearling at oneyear intervals. Kaplan-Meier curves were also drawn. Finally, we considered it possible that some variables (risk factors) would confound the effects of others. Therefore, Cox proportional hazards regression was used. The association of a variable with pearling was considered by testing its additive importance to a model for pearling that includes the possible confounding variables (i.e., the test identified the statistical significance of a single risk factor after adjusting for the other risk factors). This is an important step in analyzing the results. If the results indicate that a certain IOL feature is associated with less capsular pearling, we are unable to conclude that this is true unless we have an ideal situation in which the clinical circumstances of each factor are exactly the same or unless the confounding statistics are applied to include the influence of other clinical circumstances (risk factors). The estimated probability of pearling with each risk factor after four years of follow-up was compared to the actual numbers that developed pearling. Because we were dealing with censored data, the statistical significance of any risk factor to cause pearling was calculated from the estimated probability and not from the actual number that developed pearling.

Patients analyzed in this study had cataract surgery between April 1, 1986, and March 31, 1988. They were followed for the development of capsular pearling from April 8, 1986, to April 8, 1990. Procedures were performed by L.S. (referred to as surgeon S) or one of a group of residents (referred to as surgeon R). From the medical records we collected information on the patient's age at time of surgery and at laser caps ulotomy, the patient's sex, the surgeon, the surgical technique, the date of surgery, surgical complications, type oflens used, development of capsular pearling, and the date of laser capsulotomy. Patients were arbitrarily divided into two groups-50 years or under and over 50 years. To be included in the study cases had to have uncomplicated surgery, a plano-convex posterior chamber IOL (convex side inserted anteriorly) with a laser ridge and angled haptics, and follow-up for a minimum of two years and a maximum offour years. All age groups were included. We excluded cases with complicated intraoperative outcome (intraoperative inadvertent capsulotomy with or without vitreous loss), cases that had combined pro- Table I. Patients' characteristics. cedures (i.e., ECCE, posterior chamber IOL implantaCharacteristic tion, and trabeculectomy), cases with known glaucoma or that had argon laser trabeculoplasty or other intra- Patients 50 years of age or under ocular surgical procedures after cataract extraction, and Patients over 50 years of age cases with a history of chronic uveitis. We also excluded cases with only posterior capsule fibrosis, even if laser Mean age of patients at surgery capsulotomy was required, because the mechanism for Mean age of patients at capsulotomy postcapsular fibrosis is different from that for epithelial Females pearling. All patients had one of three types of ECCE with Males posterior chamber IOL implantation using the endocapsular (envelope) technique. Cortical material was re- Standard ECCE moved by automated aspiration/irrigation. Most Phacoem ulsification patients had standard ECCE with manual expression of Lens aspiration ECCE the nucleus; some had phacoemulsification or soft lens aspiration (Table 1). All patients had a posterior scleral Cases done by surgeon S pocket incision combined with a continuous monofil- Cases done by surgeon R ament polyester fiber suture closure (10-0 Mersilene™). 6 mm optic Posterior chamber IOLs used in this study included IOLAB 706F, 707G, 756B, 756E, G757B, 706G, and 7 mm optic G757G. 13.5 mm IOL The indications for capsulotomy were (1) subjective decrease or distortion of vision together with slitlamp 14.0 mm IOL 334

J CATARACT

R~FRACT

SURG-VOL 19, MAY 1993

Value 59 (11 %) 483 (89%) 68 years 70 years 311 (57%) 231 (43%) 500 (92%) 26 (5%) 16 (3%) 492 (92%) 50 (8%) 83 (15%) 459 (85%) 354 (65%) 188 (35%)

DISCUSSION

Table 2. Estimated probability of pearling and the actual incidence of pearling. Duration of Follow-up in Years

Kaplan-Meier Probability of Pearling (%)

Actual Cases with Pearling (%)

5

4

2

15

13

3

24

18

3.5-4.0

29

21

I

RESULTS Of the 602 consecutive cases reviewed, 542 fulfilled the inclusion criteria. Five hundred forty-two cases were followed for at least two years; 376 cases, for at least three years; 126 cases, from 3.5 to 4.0 years. Age, sex, surgical technique, and other characteristics are shown in Table 1. The estimated probability of pearling and the actual number of cases that developed pearling requiring laser capsulotomy during the four years of follow-up are shown in Table 2. Table 3 shows the estimated probability of pearling by age, sex, optic size, IOL size, surgeon, and surgical technique. When the difference between the groups within each of these risk factors was adjusted for confounding, only younger age (50 years and under) was statistically significant (P = .003). The other risk factor that remained close to statistical significance (borderline) was the larger optic (P = .54).

In the literature many factors have been implicated in the increased frequency of posterior capsule opacification, including plano-convex optic shape design,20,2! lens decentration,22 presence of IOL loops outside the capsular bag,23 use of an IOL with uniplanar haptics,! and three-piece 10L. 24 The mere presence of a posterior chamber IOL of any design or shape seems to be associated with a lower incidence of posterior capsule opacification than no implantation of an IOL.23,25 Other factors associated with less pearling include poly(methyl methacrylate) lenses,24,26 biconvex design,24 convex posterior design, !,27 an IOL with angulated haptics, ! and presence of a laser ridge. 28 ,29 Many experimental intraoperative techniques and irrigating substances tested in animals have been reported to reduce pearling; these include the use of intraocular sodium hyaluronate along with a cryo application to the posterior capsule. 30 Antiproliferative compounds such as daunomycin have been shown to reduce the proliferative capability of porcine lens epithelium. 3! It is not known whether it has the same effect on human lens epithelium. In our series of 542 cases of uncomplicated ECCE and plano-convex posterior chamber IOL implantation, 21 % of the cases actually developed pearling requiring capsulotomy; the estimated number that would have developed pearling, using the Kaplan-Meier curves, was 29%. This figure is in the middle of the previously reported range-8 % to 50%. This wide range resulted from mUltiple factors, some of which are avoidable while others are not. For example, most studies, including

Table 3. Estimated probability of pearling and the actual number of cases that developed pearling by risk factor after four years of follow-up.

Risk Factor

Kaplan-Meier Probability of Pearling Statistical Significance Between Groups % Before Adjustment for Confounding n

years

22

(37)

>50 years

106

(22)

Females

68

(22)

Males

55

(24)

6 mm optic

13

( 16)

7 mm optic

147

(32)

13.5 mm IOL

106

(30)

30

(16)

~50

14.0 mm IOL

Actual Cases with Pearling %

p= .03

32.0 18.0

p= >.064

20.9 20.3

p= .002

13.0 24.0

p= .006

25.0 16.0

p= >.030

21.0 16.0

P = .009*

20.0

Surgeon S cases

148

(30)

Surgeon R cases

11

(21)

130

(26)

Phacoem ulsification

8

(31)

19.0

Aspiration

9

(56)

50.0

ECCE

* Significance of difference between the aspiration and phacoemulsification groups and the standard ECCE group. J CATARACT REFRACT SURG-VOL 19, MAY 1993

335

ours, are retrospective, often involving more than one overall capsulotomy rate of 20% for their patients 40 kind of IOL, more than one operating surgeon, and years or under and 7% for those who were between 66 more than one surgical technique. The postoperative and 69 years. In our study, younger age was the only follow-up usually varies. Most studies underscore the statistically significant risk factor for developing pearling age of the patients studied. The average age and ethnic after confounding statistics were applied. This is a critgroup of the patients could be different in each study ical point and indicates the importance of including suggesting that we are dealing with different popula- patient age when analyzing the significance of risk factions. The indications for laser capsulotomy are often tors in the development of capsular pearling. Young different in individual studies depending on the sur- patients may have a greater number of available epigeon's criteria. Lumping pearling together with capsular thelial cells that demonstrate a greater potential to profibrosis, as well as mixing complicated and uncompli- liferate and migrate. Although this assumption is cated surgical cases, leads to variability in the reported appealing in the clinical sense, it is not supported by pathologic studies. 33 Another possible explanation is incidence of capsule opacification. In the present study the majority of cases were done that epithelial cells in the young are more juvenile and by one surgeon, complicated cases were excluded, the therefore less affected by "toxic substances" and meage of the patients was considered, all patients had one chanical inhibition caused by contact with the IOL. As was expected, the sex of the patient was not a type ofIOL (plano-convex, angled haptics, laser ridge), and only cases with epithelial pearling were considered. significant risk factor. The smaller plano-convex optic was associated with Cases with capsular fibrosis were excluded. We believe that capsular fibrosis is a secondary event caused by less pearling than the larger optic. Previous reports20 ,27 inflammation, the presence ofthe IOL, or both, that can have indicated that intimate contact between a biconvex prevent cellular migration and induce fibrosis. We sup- IOL and the posterior capsule would decrease the inciport our beliefs with the following observations: (1) Ster- dence of pearling; thus, one would expect the smaller ling and Wood 27 studied the effect ofIOL convexity on plano-convex optic to be associated with more intimate posterior capsule opacity and noted that with the convex contact and less pearling, and this was what we found. posterior IOL a smooth, fibrous type of capsule opacity The IOL optic was the only risk factor with a borderline prevailed, whereas with the convex anterior IOL pear- significance (P = .54) after confounding statistics were ling was the prevailing type of posterior capsule opacity. applied. Our series also showed that IOLs with a smaller (2) Certain clinical reports indicate that a posterior overall diameter were more likely to be associated with chamber IOL retards pearling and not fibrosis.2o (3) The pearling than lenses with a larger overall diameter. This in vitro studies by Santo and coauthors26 have demon- was not statistically significant when the confounding strated that an increase in the capsule-optic surface con- effect of other risk factors was considered. It is possible tact may contribute to fibrosis. While this contact with that the larger overall size provides better centration and the IOL inhibits lens epithelial migration, some epi- therefore less chance of pearling. thelial cells that escape beneath the optic undergo pleoOur study showed that the probability of developing morphism, suggesting myofibroblastic metaplasia. pearling requiring laser capsulotomy was 56% after lens Nishi,23 Frezzoti and Caporossi,25 and Sellman and aspiration ECCE. Although this was significantly higher Lindstrom 2o consider pearling and fibrosis as intrinsic than the probability in the standard ECCE and phacophenomena of the posterior capsule. emulsification groups, we do not think it is a true difAnother important factor in the development of pear- ference. All the patients who had aspiration were in the ling is the time since the surgery. Only 4% of our patients younger age group (50 or under), which is the group that required capsulotomy after one year, whereas 13% and had the most significant risk of developing pearling re18% required it after two and three years, respectively. quiring capsulotomy. Prospective studies in rabbits 34 This agrees with other reports. 1,32 The delay is probably have shown that the type of surgical technique used in due to multiple factors, including slow migration and cataract extraction has no effect on the incidence of proliferation of epithelial cells, early mechanical or bio- posterior capsule opacification. Furthermore this differchemical inhibition (or both) of cellular migration ence was not statistically significant after we applied caused by the presence of an IOL, which is later over- confounding statistics. This finding again illustrates the come by the migrating lens epithelium. need to consider the patient's age as the single most Cataracts are uncommon in patients under 40 years important risk factor for the development of pearling, as of age. If cataracts occur in this age group there is usually well as to apply confounding statistics when dealing with a familial history, or they are related to trauma, second- multifactorial problems. ary to intraocular disease, or secondary to systemic or The surgeon's skill did not seem to affect the incitopical medications. Our data analysis indicated that dence of pearling, which occurred with approximately patients younger than 50 have a two-fold higher risk of the same frequency in patients operated on by surgeon developing pearling than do those who are older than 50. S and surgeon R. The importance of the surgeon's skill Frezzotti and Caporossi,25 in a review of 223 cases of might have been underestimated in this study for two ECCE with and without IOL implantation, reported an reasons. First, surgeon R (resident group) operated on 336

J CATARACT REFRACT SURG-VOL 19. MAY 1993

fewer cases. Second, procedures by surgeon R were under direct supervision of surgeon S and therefore to a degree used the skills of surgeon S. Factors that might influence the incidence ofpearling include the degree to which the surgeon polished the posterior capsule and whether the capsular bag was entirely cleaned of all cortical remnants. The surgeon's skill may be a more important factor in inducing capsular fibrosis than in inducing pearling. All 10Ls used in this study had angled haptics and a laser ridge. Angled haptics have been reported to have a lower incidence of pearling. 1 This has been attributed to more intimate contact between the lens and the posterior capsule because of the angulation. The laser ridge creates a space between the lens and the posterior capsule. It was originally conceived to reduce the risk ofIOL damage when the patient had laser capsulotomy. There is no consensus on whether a laser ridge can reduce or increase the incidence of pearling. Hansen and coauthors29 reported that after an 18 month follow-up, the cumulative capsulotomy rate in eyes with planoconvex lenses was 7.1 % and in eyes with a laser ridge it was 1.8%. Maltzman and coauthors 28 showed a clinical benefit from using the laser ridge that was not statistically significant. On the other hand, in a retrospective study of 1,040 cases, Born and Ryan 21 reported a cumulative capsulotomy rate of 40% in patients who had laser ridge lenses.

CONCLUSION Our retrospective study differs from other studies in that the majority of procedures were performed by a single surgeon using similar 10L styles. Twenty-one percent of patients who had uncomplicated ECCE (envelope-endocapsular technique) with implantation of an angled plano-convex laser ridge posterior chamber IOL developed capsular pearling that required laser capsulotomy after a four year follow-up. Age of the patient was the most important risk factor for the development of posterior capsular pearling. We recommend that future studies of the incidence of capsular opacification do the following: (I) make a distinction between pearling and fibrosis; (2) consider only cases of pearling in data analysis; (3) consider only uncomplicated cases; (4) apply confounding statistics to offset the confounding effect of risk factors that are difficult to standardize.

REFERENCES 1. Downing JE. Long-term discission rate after placing posterior chamber lenses with the convex surface posterior. 1 Cataract Refract Surg 1986; 12:651-654 2. Lindstrom RL, Harris WS. Management of the posterior capsule following posterior chamber lens implantation. Am Intra-Ocular Implant Soc 1 1980; 6:255-258 3. Sinskey RM, Cain W lr. The posterior capsule and phacoemulsification. Am Intra-Ocular Implant Soc 1 1978; 4: 206-207

4. Wilhelmus KR, Emery 1M. Posterior capsule opacification following phacoemulsification. Ophthalmic Surg 1980; 11 :264-267 5. Bath PE, Fankhauser F. Long-term results of Nd:YAG laser posterior capsulotomy with the Swiss laser. 1 Cataract Refract Surg 1986; 12:150-153 6. Ficker LA, Steele ADM. Complications ofNd:YAG laser posterior capsulotomy. Trans Ophthalmol Soc UK 1985; lO4:529-532 7. Derner lL, Koch DD, Smith lA, Knolle GE lr. Persistent elevation in intraocular pressure after Nd: YAG laser treatment. Ophthalmic Surg 1986; 17:465-466 8. Channell MM, Beckman H. Intraocular pressure changes after neodymium-YAG laser posterior capsulotomy. Arch Ophthalmol 1984; lO2: lO24-lO26 9. Dardenne M-U, Gerten G-l, Kokkas K, Kermani O. Retrospective study of retinal detachment following neodymium:YAG laser posterior capsulotomy. 1 Cataract Refract Surg 1989; 15:676-680 lO. Rickman-Barger L, Florine CW, Larson RS, Lindstrom RL. Retinal detachment after neodymium:YAG laser posterior capsulotomy. Am 1 Ophthalmol1989; lO7:531536 It. Neuteboom, GHG, De Vries-Knoppert WAE1. Endophthalmitis after Nd: YAG laser capsulotomy. Doc Ophthalmol 1988; 70:175-178 12. Carlson AN, Koch DD. Endophthalmitis following Nd: YAG laser posterior capsulotomy. Ophthalmic Surg 1988; 19:168-170 13. Tetz MR, Apple OJ, Price FW lr, et al. A newly described complication of neodymium-YAG laser capsulotomy: exacerbation of an intraocular infection. Arch Ophthalmol 1987; lO5:1324-1325 14. Meisler DM, Zakov ZN, Bruner WE, et al. Endophthalmitis associated with sequestered intraocular Propionibacterium acnes. Am 1 Ophthalmol 1987; lO4:428 15. Blacharski PA, Newsome DA. Bilateral macular holes after Nd:YAG laser posterior capsulotomy (letter). Am 1 Ophthalmol1988; lO5:417-418 16. Schaeffer AR, Ryll DL, O'Donnell FE Jr. Ciliochoroidal effusions after neodymium:YAG posterior capsulotomy: association with pre-existing glaucoma and uveitis. 1 Cataract Refract Surg 1989; 15:567-569 17. Weinreb RN, Wasserstrom lP, Parker W. Neovascular glaucoma following neodymium-YAG laser posterior capsulotomy. Arch Ophthalmol 1986; lO4:730-731 18. Paylor R. Central retinal artery occlusion following YAG synechialysis (letter). Arch Ophthalmol 1985; lO3:325326 19. Kalbfleish lD, Prentice RL. The Statistical Analysis of Failure Time Data. New York, 10hn Wiley and Sons, 1981 20. Sellman TR, Lindstrom RL. Effect of a plano-convex posterior chamber lens on capsular opacification from Eischnig pearl formation. 1 Cataract Refract Surg 1988; 14:68-72 21. Born CP, Ryan DK. Effect of intraocular lens optic design on posterior capsular opacification. 1 Cataract Refract Surg 1990; 16: 188-192 22. Tetz MR, O'Morchoe OJC, Gwin TD, et al. Posterior capsular opacification and intraocular lens decentration. Part II: Experimental findings on a prototype circular intraocular lens design. 1 Cataract Refract Surg 1988; 14: 614-623

JCATARACfREFRACfSURG-VOL 19, MAY 1993

337

23. Nishi O. Incidence of posterior capsule opacification in eyes with and without posterior chamber intraocular lenses. J Cataract Refract Surg 1986; 12:519-522 24. Hansen SO, Solomon KD, McKnight GT, et al. Posterior capsular opacification and intraocular lens decentration. Part I: Comparison of various posterior chamber lens designs implanted in the rabbit model. J Cataract Refract Surg 1988; 14:605-613 25. Frezzotti R, Caporossi A. Pathogenesis of posterior capsular opacification. Part 1. Epidemiological and dinicostatistical data. J Cataract Refract Surg 1990; 16:347-352 26. Santos BA, Pastora R, DelMonte MA, O'Donnell FE Jr. Lens epithelial inhibition by PMMA optic: Implications for lens design. J Cataract Refract Surg 1986; 12:23-26 27. Sterling S, Wood TO. Effect of intraocular lens convexity on posterior capsule opacification. JCataract Refract Surg 1986; 12:655-657 28. Maltzman BA, Haupt ED, Cucci P. Effect of the laser ridge on posterior capsule opacification. J Cataract Refract Surg 1989; 15:644-646

338

29. Hansen TE, Otland N, Corydon L. Posterior capsule fibrosis and intraocular lens design. JCataract Refract Surg 1988; 14:383-386 30. Zaturinsky B, Naveh N, Saks D, Solomon AS. Prevention of posterior capsular opacification by cryolysis and the use of heparinized irrigating solution during extracapsular lens extraction in rabbits. Ophthalmic Surg 1990; 21 :431434 31. Weller M, Wiedemann P, Fischbach R, Hartmann C, Heimann K. Evaluation of daunomycin toxicity on lens epithelium in vitro. Int Ophthalmol 1988; 12:127-130 32. Liesegang TJ, Bourne WM, I1strup DM. Secondary surgical and Nd: YAG laser discissions. Am J Ophthalmol 1985; 100:510-519 33. Argento C, Zarate J. Study of the lens epithelial cell density in cataractous eyes operated on with extracapsular and intercapsular techniques. J Cataract Refract Surg 1990; 16:207-210 34. Odrich MG, Hall SJ, Worgul BY, et al. Posterior capsule opacification: experimental analyses. Ophthalmic Res 1985; 17:75-84

J CATARACT REFRACTSURG-VOL 19. MAY 1993