Calculating an IOL power in an adult is simple: the surgeon chooses a power to give emmetropia based on measurements of axial length (AL) and cornea power (K) using one of the formulas designed for adult eyes, such as the Hoffer Q, Haigis, Holladay, or SRK-T formula; in published series they show a mean absolute error of <0.6 D (diopter).¹ In contrast, the calculation of IOL power in children is complicated by less accurate biometry, the larger effect of measurement error in small eyes, and possibly by the lack of a published IOL formula specifically for children.

Studies of IOL implants in children demonstrate larger formula prediction errors than are found when the same formulas are used in adults. Mezer et al.² found that the mean difference between the predicted and the actual postoperative refractions was slightly more accurate using theoretic formulas (1.06 D versus 1.22 D with regression formulas). Andreo et al.³ studied 47 pediatric patients, age 0.25 to 16 years. They measured the initial postoperative pseudophakic refractions and compared them to those predicted by four formulas (SRK-II, SRK-T, Holladay, and Hoffer Q). They found that the average initial postoperative refractive error was between 1.2 and 1.4 D. Moore et al.⁴ performed a retrospective review of 50 pediatric eyes undergoing secondary IOL implantation, and found a mean absolute value prediction error of 1.64 D.

Biometry (measurement of AL and K) in young children is typically performed in the operating room with the child asleep under general anesthesia, and is usually less precise than in adults. The nonstandard conditions for biometry limit the tools to those that are portable to the operating room; some instruments used for biometry in adults such as the IOLMaster cannot be used. The AL may be difficult to measure accurately because the anesthetized child cannot voluntarily align his or her line of sight with the axis of measurement; in addition, a small amount of pressure from the ultrasonic A-scan transducer can easily deform a child’s soft cornea. The cornea power is also difficult to measure accurately: artifacts induced by drying or by pressure on the eye from a speculum can distort the cornea curvature.

Errors of biometry in small eyes lead to larger errors in IOL power calculation than do the same errors in adults. This is because the effect of these errors is inversely related to the size of the eye: a 0.1-mm error of measurement in a 24-mm eye is an error of 0.4%; the same error in a 16-mm eye is an error of 0.6%. We calculate that this 0.1 mm error would result in an absolute refractive error of 0.23 D in the 24-mm eye and 0.57 D in the 16-mm eye (Table 7.1). Thus biometric errors are propagated by IOL calculation formulas and magnified in small eyes.

All modern IOL calculation formulas are based on adult eyes. They assume that the IOL will rest in its usual position in the adult-sized eye, in relation to AL, cornea curvature, and (in some formulas) other parameters. In addition to their basis in optics theory, some of the formulas are empirically modified to optimize results in adult eyes. It is unknown precisely how much children’s eyes vary from adult proportions. Some of the observed errors in the calculation of initial postoperative refractions after pediatric IOL implantation may be due to incorrect assumptions in these formulas or to unusual proportions in pediatric eyes with some types of cataract. However, most children’s eyes eventually grow to adult proportions, so adult formulas may be better in the end than a formula designed for pediatric eyes.

The combination of increased errors of biometry, the greater effect of these errors on IOL calculations, and the lack of an IOL formula for children may be the primary cause of the observed lesser accuracy of IOL calculations in children than in adults. In addition, postoperative refractions are more difficult to obtain precisely in children than in adults, and there may be a small growth of the eye between IOL implantation and the first refraction; these additional reasons may help to explain why published studies find a greater error in pediatric eyes. Despite these causes, a carefully done recent study by Trivedi et al.⁵ of 45 eyes with surgery at a mean age of 3.56 years found a remarkably low mean absolute error of 0.68 to 0.84 D, using four theoretic IOL formulas; the Holladay 2 formula gave slightly better predictions than the others. The authors of this study clearly used careful measurement techniques including immersion ultrasound for AL; these results may be close to the best possible in theory.

High refractive error in unilateral aphakia makes it very difficult to obtain good vision in the long term due to amblyopia. Children with unilateral aphakia are generally treated with contact lenses that are not always worn continually; the contact lens can also cause blur when it is displaced to the edge of the pupil in daily life. Even short intervals of uncorrected aphakia can cause dense and long-term amblyopia.⁶ The hope of IOL implantation is that a constantly sharp retinal image will simplify amblyopia management and result in better vision. A large long-term retrospective study found a better mean visual acuity for pseudophakic eyes than aphakic eyes in all age groups (Table 2 in McClatchey et al.⁷). A small study of infants who received a primary IOL after extraction of dense congenital unilateral cataract showed improvement from a mean visual acuity of 20/170 at 6 months to 20/85 at 12 months and 20/54 at 4 years. Visual acuity in the IOL group at 4 years was similar to that of children who had good-to-excellent contact lens wear compliance and better than that of children who had moderate-to-poor compliance.⁸

A survey of the American Association for Pediatric Ophthalmology and Strabismus (AAPOS) in 2001 found that most members preferred to leave infants aphakic after cataract surgery and use contact lenses for initial correction.⁹ It is certainly possible to achieve good vision with this strategy: two of our patients with unilateral cataract surgery who were left aphakic have achieved 20/25 vision, thanks to constant use of contact lenses and consistent patching.

The Infantile Aphakia Treatment Study is a randomized, multicenter clinical trial of 114 infants with a unilateral congenital cataract comparing IOL implantation at cataract surgery to leaving the infant aphakic. Preliminary results of logMAR grating visual acuity at 1 year of age failed to show a significant difference between the treated eyes in the two groups (contact lens group, 0.80; IOL group, 0.97; P = 0.19).¹⁰ In addition, there were more patients with intraoperative complications (28% versus 11%; P = 0.03), adverse events (77% versus 25%; P < 0.0001), and additional intraocular surgeries (63% versus 12%; P < 0.0001) in the IOL group than the contact lens group.¹¹

Thus the evidence on whether IOL implantation improves ultimate amblyopia management of children with unilateral cataract surgery is incomplete, and age may be a factor. Certainly IOL implantation is not sufficient by itself in most children, and the surgeon must plan to manage amblyopia using standard treatment techniques: patching and correction of residual refractive error. Frequent and diligent follow-up must be vigorously pursued, and parents must be taught the importance of patching, often repeatedly.

Anisometropia is of great concern for children who have unilateral cataract surgery; bilateral cataract surgery generally results in nearly equal refractions both initially and
with increasing age, though the two eyes of a child may grow at different rates in some. The problems due to the often substantial anisometropia in children with unilateral pseudophakia are best broken down into several categories:

Therefore, the clinician who manages children with unilateral pseudophakia must be prepared to manage these several aspects of anisometropia. In the case of a child who wears spectacles, this becomes significant when the anisometropia is greater than about 4 D. Depending on the age, cooperation with treatment, and ease of contact wear, this can be managed by switching from spectacles to contact lenses. In some, reducing the power of the minus spectacle lens in the pseudophakic eye by about a diopter will allow good near vision with that eye while reducing the aniseikonia and anisophoria to acceptable levels.