Changes in Astigmatism After Congenital Cataract Surgery and IntraocularLens Implantation: A Comparative Study

Objective  To evaluate the postoperative changes in astigmatism in the pseudophakiceyes of children who underwent 1 of 3 different types of surgical incisionsfor congenital cataract extraction with intraocular lens implantation, andin whom astigmatism of at least 3 diopters (D) was recorded 1 week after theoperation.

Methods  We retrospectively reviewed the medical records of all the childrenin our department who had undergone surgery for nontraumatic cataract between1992 and 2001. Cataract surgery with intraocular lens implantation was performedusing 1 of 3 types of surgical incisions: a limbal incision, a scleral tunnel,or a clear corneal incision allowing the use of a foldable intraocular lens.In 28 children (32 eyes) aged 2 months to 11 years (mean ± SD, 4.7± 3.4 years), astigmatism of 3 D or more was found when assessed 1week after surgery. The refraction was measured and recorded again 3 monthsand 5 months after surgery. The paired t test wasused to compare the outcome variables.

Main Outcome Measures  Refractive error 1 week, 3 months, and 5 months after surgery.

Results  Mean ± SD astigmatism 1 week postoperatively was 5.8 ±2.2 D, 5.1 ± 2.1 D, and 4.0 ± 1.3 D in groups 1, 2, and 3, respectively.Thereafter, the astigmatic component of the refractive error underwent a spontaneousdecline, reaching mean ± SD values of 0.9 ± 1.0 D, 1.6 ±1.6 D, and 1.0 ± 0.8 D, respectively, in the 3 groups 5 months afterthe operation. The difference between the mean values at 1 week and at 5 monthsin each group was statistically significant (P <.001 in group 1; P = .01 in group 2; and P<.001 in group 3).

Conclusion  Children who underwent extraction of congenital cataract and intraocularlens implantation by different surgical techniques showed a significant spontaneousreduction in astigmatism postoperatively.

Corneal astigmatism after cataract surgery is a well-documented findingin adults. The amount of astigmatism depends on various factors, such as thetype and location of the surgical incision, the amount of scleral cauterizationperformed, the suturing material used, suture placement, and postoperativeuse of steroids.^1-3 Inchildren as well as in adults, a preoperative existing astigmatism also influencesthe postoperative refractive error.

In adults, only mild spontaneous changes in the amount of postoperativeastigmatism have been described.^4,5 Aneffective way to reduce or eliminate the postoperative astigmatism is throughthe removal of 1 or more interrupted or continuous sutures.^6-9 Thisprocedure relieves wound compression, thereby altering the corneal curvature.Suture removal is recommended only in eyes with postoperative astigmatismof at least 3 diopters (D).^7,10

Recent reports have documented the finding of spontaneous regressionin astigmatism after cataract surgery in children.^11,12 Thesestudies reported the changes in astigmatism that occurred without suture removalafter cataract surgery and intraocular lens (IOL) implantation in childrenwith cataract.

The medical records of all 73 children (112 eyes) who underwent surgery for congenital cataract between 1992 and 2001were reviewed. All eyes found to have astigmatism of at least 3 D one weekafter surgery were included in the study. There were no other ocular or systemicabnormalities in the study population.

All patients had undergone extracapsular cataract extraction and IOLimplantation. Three types of surgical incisions were used. The following procedurewas common to all techniques. Two paracentesis ports were opened at the limbusat the 2- and 10-o'clock positions. An anterior chamber maintainer (Visitec,Warwickshire, England) was used, and capsulorrhexis was performed with a bent25-gauge needle. Lens material was aspirated with an aspirating cannula (Anis;Storz, St Louis, Mo). In some eyes, posterior capsulotomy and anterior vitrectomywere performed with a vitrector instrument (Occutome; CooperVision, Irvine,Calif). The corneal incision was closed with interrupted 10-0 sutures (Mersilene;Ethicon, Edinburgh, Scotland).

In group 1 (27 children, 45 eyes), the following modifications weremade. A fornix-based conjunctival flap was formed and a partial-thicknessincision was made at the limbus with a No. 64 Beaver blade (Sable Industries,Oceanside, Calif). The corneal incision was enlarged with scissors. In group2 (24 children, 34 eyes), we modified the method by creating a scleral groove2 mm from the limbus with a No. 64 Beaver blade. A scleral pocket was constructedwith a crescent knife, and dissection was extended for 1 mm into the clearcornea. Using a 3.6-mm keratome (Visitec), the anterior chamber was enteredthrough the scleral tunnel. In group 3 (22 children, 33 eyes), the corneawas entered at the 12-o'clock meridian just anterior to the terminal endsof the conjunctival blood vessels as they cross the limbus. A balance-typeIOL (Hanita Lenses, Kibbutz Hanita, Israel) was implanted in the eyes of groups1 and 2, and a foldable IOL (AcrySof, Alcon, Tex) was used in group 3. Inmost cases, the IOL was implanted in the capsular bag, but in some cases,it was placed in the sulcus.

All operations were performed by the same surgeon (A.S.), using thesame surgical technique modified as described for each group. All patientswere treated postoperatively with dexamethasone sodium phosphate and neomycinsulfate eye drops applied 6 times a day for 1 week and then 4 times a dayfor an additional 2 weeks and with 0.5% tropicamide twice a day for 2 weeks.

The refractive error of the surgically treated eye was measured (aspart of a complete eye examination) with streak retinoscope 1 week after surgeryand then, on average, every 1 to 2 months for 5 months after dilation of thepupil with 0.5% tropicamide. In all patients, the refractive error was correctedwith spectacles 1 month after surgery, and treatment for amblyopia was institutedwhen indicated. No sutures were removed during follow-up. The paired t test was used to compare the outcome variables at thedifferent postoperative periods.

When examined 1 week after surgery, 32 eyes (28 children [14 boys and14 girls]) had astigmatic errors of 3 D or more and were therefore includedin the study. In 4 patients (1 patient each in groups 1 and 3, and 2 patientsin group 2), both eyes were operated on with the same technique. In thesecases, the 2 eyes were averaged for the analyses. Thus, 28 cases were includedin the study. The children were aged 2 months to 11 years (mean ± SD,4.7 ± 3.4 years). Mean ± SD ages in the 3 groups were 6.8 ±3.6, 3.2 ± 2.8, and 3.5 ± 2.7 years, respectively. In each case,both the surgical and the postoperative course were uneventful, and therewere no complications.

Table 1 presents the meanvalues and ranges of astigmatism after surgery in the different groups. Theeyes in all 3 groups showed a decline in mean astigmatism with time. In all3 groups, the mean change in astigmatism was greater during the first 3 postoperativemonths than during the following 2 months (Table 1).

The observed changes in astigmatism between 1 week and 5 months aftersurgery were statistically significant in all 3 groups (P<.001 in group 1; P = .01 in group 2;and P <.001 in group 3).

The immediate postoperative astigmatism and its subsequent changes areaffected by factors such as the surgical technique, type of suture used, andexperience of the surgeon.^1-3,13 Cataractsurgery in children, with or without anterior vitrectomy, can be done throughsmall limbal incisions.¹⁴ For IOL implantation,the incision must be enlarged in accordance with the diameter of the lensto be implanted. Postoperative astigmatism may be affected by the width ofthe incision constructed in the sclera and the need to suture this cut inchildren. The relatively small astigmatism reported after scleral tunnel inother studies may have been attributable to sutureless surgery.^15-17

In the present study, postoperative stigmatism showed a spontaneousdecrease during the period between 1 week and 5 months after surgery, withmost of this decrease seen during the first 3 months. The decline in astigmatismoccurred in all patients regardless of the mode of surgical incision. Astigmatismdeclined in groups 1, 2, and 3 from 5.8 D, 5.1 D, and 4.0 D, respectively,1 week after surgery to 0.9 D, 1.6 D, and 1.0 D, respectively, 5 months aftersurgery. Brown et al¹² recently reported spontaneousrelaxation of postoperative astigmatism in children after lens implantationthrough a 6.25-mm scleral wound. In that study, the mean astigmatism was 6.71D at 1 to 15 days after surgery and 1.93 D at 31 to 45 days. The authors concludedthat surgeons should not hesitate to secure scleral wounds meticulously inchildren because of fear of a permanent undesirable refractive outcome.

In our patients, the astigmatism decreased more rapidly, on average,during the first 3 months after the operation than during the following 2months. This decline in astigmatism with time was observed in all 3 groups.Similar findings were reported in adults,^7,18 inwhom surgery-induced astigmatism declined rapidly during the first 1 or 2months postoperatively. In another study,¹⁹ thecorneal curvature stabilized at 19 weeks after surgery. On the basis of theirown results, Talamo et al⁷ concluded that itmay be unwise to cut the sutures in patients with early postoperative astigmatismof less than 3 D because of the instability of the corneal astigmatism.

The finding of postoperative astigmatism is important, especially inchildren, because of its adverse effect on vision development and the riskof amblyopia.²⁰ In adults, a few months ofdelay in correcting the refractive error does not affect the final visualacuity, whereas in children, the optical refraction must be precisely correctedas soon as possible. Accordingly, the refractive error in all of our patientswas corrected 1 month after surgery, and any further changes in astigmatismwere promptly attended to by changes in the refractive correction. On theother hand, whereas postoperative surgical astigmatism in adults can be correctedby removal of 1 or more sutures in a simple office procedure,^6,7 inchildren this procedure usually requires general anesthesia.

Mild spontaneous regression of postoperative astigmatism has been describedin adults. Without suture cutting, mean changes of only 0.5 D⁴ and1.25 D⁵ were reported during the first yearafter cataract surgery. In our present series of pediatric patients, by 5months after surgery there was a spontaneous average decline in astigmatismof 4.9 D, 3.5 D, and 3.0 D in groups 1, 2, and 3, respectively (Table 1). At least 2 factors might have contributed to this markedspontaneous regression. First, the ocular tissues in children exhibit a highdegree of elasticity. In adults, wound compression caused by the sutures doesnot change across time, whereas in children, because of the elasticity ofthe cornea and sclera, the tissue tension may spread evenly to neighboringareas and reduce the amount of astigmatism. Second, growth of the globe inchildren (but not in adults) continues under constant centrifugal intraocularpressure and results in a more spherical growth of the eye, thereby diminishingthe amount of astigmatism. This factor may be particularly important in youngchildren.

In adults, suture removal is recommended as a way to correct postoperativeastigmatism. The findings of the present study strongly suggest that removalof sutures in children is not required, because the astigmatism regressesspontaneously a few months after surgery. This would eliminate the need forthe general anesthesia that is usually necessary for suture removal in children.

Corresponding author: Abraham Spierer, MD, Goldschleger Eye Institute,Sheba Medical Center, Tel-Hashomer 52621, Israel (e-mail: spierera@post.tau.ac.il).

Submitted for publication April 3, 2003; final revision received August12, 2003; accepted August 25, 2003.

This study was supported by a grant from the Sackler Faculty of Medicine,Tel-Aviv University, Tel-Aviv, Israel.