To compare the refractive outcome and postoperative complications of cataract surgery among nanophthalmos and relative anterior microphthalmos and the normal control eyes.
Retrospective case-control series.
Seventeen eyes with nanophthalmos, 29 eyes with relative anterior microphthalmos, and 54 normal control eyes were enrolled in this study. The subjects were divided into 3 diagnostic subgroups according to the following: nanophthalmos with an axial length <20.5 mm and without morphologic malformation; relative anterior microphthalmos with a corneal diameter (CD) ≤11 mm, an anterior chamber depth (ACD) ≤2.2 mm, and an axial length (AL) ≥ 20.5 mm; and normal control group eyes defined as an AL ≥20.5 mm with a CD >11 mm or an ACD >2.2 mm. The implanted intraocular lens (IOL) power was used to calculate the predicted postoperative refraction error according to 4 IOL power formulas: SRK II, SRK/T, Hoffer Q, and Holladay 1. With each formula, the mean numeric error and mean absolute error were calculated. At postoperative 2 months, the endothelial cell count and the complications were analyzed.
As measured by mean numeric error or mean absolute error, there was a significant difference among the 3 groups based on SRK II, SRK/T, and Hoffer Q, with less predictability in the nanophthalmic eyes. In eyes with nanophthalmos, the Holladay 1 produced the best refractive results as measured by mean numeric error ( P < .001). A higher occurrence rate of posterior capsule rupture (11.7%) was shown in the nanophthalmic eyes. The difference among the 3 groups for the postoperative endothelial cell loss was not significant ( P = .421).
The refractive predictability and postoperative outcome was poorer in the eyes with nanophthalmos compared to the eyes with relative anterior microphthalmos or normal control.
The clinical spectrum of microphthalmos encompasses heterogenous “small-eye” phenotypes that vary from the extreme of anophthalmos to simple microphthalmos, pure microphthalmos or nanophthalmos, partial microphthalmos, posterior microphthalmos, and relative anterior microphthalmos. Among the spectrum of small adult eyes, nanophthalmos is characterized by a short axial length, a smaller anterior segment, and a high incidence of angle-closure glaucoma and uveal effusion. Relative anterior microphthalmic eyes are different in that they have a normal axial length (AL) but a disproportionately smaller anterior segment.
Cataract surgery in small eyes is demanding, with several potential complications, but the reported surgical outcomes for small eyes have been encouraging with the advances in phacoemulsification and intraocular lens (IOL).
In a previous study, 6 eyes with axial lengths less than 19.0 mm showed a tendency toward hypermetropia compared with the predicted refraction. However, the number of eyes in that study was too small to draw a conclusion. There is currently no definite consensus for the IOL formula in small adult eyes as well as eyes with relative anterior microphthalmos, even though some reports have suggested that the SRK/T, Hoffer Q, or Holladay II formula performed well for short eyes.
We thought it might be significant to evaluate refractive outcomes and postoperative complications of cataract surgery in small eyes and to compare postoperative results among eyes with nanophthalmos, relative anterior microphthalmos, and normal eyes.
The eyes that met the eligibility criteria were enrolled from a database of patients who underwent routine cataract surgery at our hospital between March 1, 2007 and March 31, 2010.
Nanophthalmos in this retrospective study was defined as an axial length <20.5 mm without morphologic malformation. The inclusion criteria for relative anterior microphthalmos in this study were defined in the following manner : a corneal diameter (CD) ≤11 mm, an anterior chamber depth (ACD) ≤2.2 mm, an AL ≥20.5 mm, and no morphologic malformation. The normal control groups that matched small eyes in age were defined as an AL ≥20.5 mm with a CD >11 mm or an ACD >2.2 mm among the patients who underwent cataract surgery.
The exclusion criteria included cataract hardness greater than nuclear opalescence (NO) 5 on the LOCS III scale, ocular comorbidity and an axial length greater than 25.0 mm, a history of intraocular surgery, and follow-up period less than 2 months. The corneal exclusion criteria included corneal dystrophy and corneal scarring. The systemic exclusion criteria included diabetes, the use of oral steroidal agents, and the presence of dermatologic disease. If both eyes of 1 patient had combined procedures, then 1 of the eyes was randomly selected for the study.
All of the enrolled patients were evaluated with the following: autorefraction (Auto-Ref-Keratometer, RK-5; Canon, Tokyo, Japan), best-corrected visual acuity (BCVA), slit-lamp examination, intraocular pressure (IOP) measurement, and a fundus examination. Based on the objective autorefraction, subjective verification was made for the best-corrected visual acuity. The AL was measured by an ultrasonographic A-scan (Echograph model axis II; Quantel Medical, Clermont-Ferrand, France). Measurements were made using an immersion shell containing water; there was no direct contact of the ultrasound probe with the cornea. Computerized videokeratography (Orbscan IIz; Bausch and Lomb Inc, Rochester, New York, USA) was performed to evaluate the anterior chamber depth and the corneal diameter, and the latter was measured as the white-to-white distance on the digitally processed images. The curvature radii of the cornea were measured with a keratometer (OM-4; Topcon, Tokyo, Japan). The endothelial cell count and coefficient of variation (CV) and hexagonality were obtained by specular microscopy (NonconROBO-CA, KONAN Medical Inc, Nishinomiya, Hyogo, Japan) both preoperatively and postoperatively. Corneal edema was recorded and the uveal inflammatory response was documented in the form of flare and/or cells graded by Hogan’s criteria. The percentage of eyes with flare and cells >grade 3 by Hogan’s criteria was compared among 3 groups.
The implanted IOL power was used to calculate the predicted postoperative refraction error by 4 commercially available IOL power formulas: SRK II, SRK/T, Holladay 1, and Hoffer Q. With each formula, the mean numeric error was calculated as the actual postoperative spherical equivalent minus the predicted spherical equivalent. The postoperative mean absolute prediction error (ie, the absolute value of the mean numeric error of the postoperative prediction error) was also calculated. The refractive error was calculated 2 months postoperatively.
All the surgeries were performed by an experienced surgeon using the same technique. Preoperatively, the pupil was dilated with a combination of topical tropicamide 1.0% (Ocutropic, Samil, Korea) and phenylephrine 2.5% (Mydfrin; Alcon, Fort Worth, Texas, USA), and the eyes received topical anesthesia with 0.5% proparacaine hydrochloride (Alcaine; Alcon) and 4% lidocaine. A 3.00-mm clear corneal temporal incision was made, and 1% sodium hyaluronate (Healon; AMO, Los Angeles, California, USA) or 3% sodium hyaluronate plus 4% chondroitin sulfate (Viscoat; Alcon, Puurs, Belgium) was injected into the anterior chamber. A continuous curvilinear capsulorrhexis approximately 5.5 mm in diameter was created. Its size was intended to be slightly smaller than the IOL optic size for all the cases. After careful hydrodissection and hydrodelineation using balanced salt solution (BSS; Alcon), phacoemulsification of the nucleus was performed, and the residual cortex was cleared by an irrigation/aspiration system. The lens capsule was inflated with Healon, and the yellow-tinted IOLs (YA60BB; Hoya, Tokyo, Japan) were implanted in the capsular bag. IOL implantation was performed under the protection of an ophthalmic viscosurgical device (OVD), which was subsequently removed through aspiration. The wound was sutured in 7 eyes with nanophthalmos and there were no sutures in the relative anterior microphthalmic eyes or the normal control eyes.
Statistical analyses were performed with SPSS software (version 17.0; SPSS, Chicago, Illinois, USA). Differences among the nanophthalmic and relative anterior microphthalmic eyes and the normal control eyes were assessed by Kruskal-Wallis test for continuous parameters and with the Fisher exact test for numeric parameters. Preoperative and postoperative best-corrected visual acuity (logMAR) was compared by Wilcoxon signed rank test. Within each group, the difference in the mean numeric error and mean absolute error between the different IOL calculation formulas was analyzed using Friedman test. Post hoc analysis was done by multiple comparisons using Bonferroni correction to adjust overall significance level. A P value <.05 was considered statistically significant.
Among 1299 consecutive patients who underwent routine cataract surgery at our hospital between March 2007 and March 2010, 17 eyes (17 patients) that were found to fulfill the inclusion criteria for the definition of nanophthalmos were enrolled into the study after chart review. All nanophthalmic eyes were used in analysis of outcome and complications after cataract surgery except the refractive results for 2 nanophthalmic eyes with an IOL that was inserted into the sulcus. In the 40 eyes with relative anterior microphthalmos that we could detect by a review of the charts, 11 eyes were excluded as their follow-up period was less than 2 months, and so 29 eyes (29 patients) with relative anterior microphthalmos were finally included in the current study. Fifty-four normal control eyes that fulfilled the inclusion criteria were randomly selected using a random number table.
The demographics of the eyes with nanophthalmos, the eyes with relative anterior microphthalmos, and the normal control group eyes are shown in Table 1 . The mean age of the patients in the nanophthalmos group was 55.44 ± 15.80 years, that for the relative anterior microphthalmos group was 73.36 ± 10.06 years, and that for the normal control group was 68.96 ± 11.68 years. The nanophthalmic patients were younger than the patients of the relative anterior microphthalmos or normal control groups ( P = .001). The proportion of male subjects was 0% in the nanophthalmos group, 11.5% in the relative anterior microphthalmos group, and 74.1% in the normal control group ( P < .001). A statistically significant difference in the laterality between the 3 groups was not detected. Seven of 17 eyes (41.1%) with nanophthalmos had glaucoma and 9 of 29 eyes (31.0%) with relative anterior microphthalmos had glaucoma. Of these, 6 of 7 eyes (35.3%) in nanophthalmos group and 4 of 9 eyes (13.7%) in relative anterior microphthalmos group were undergoing antiglaucomatic medical treatment. Four of the 17 eyes (23.5%) in the nanophthalmos group and 4 of 29 eyes (13.7%) in the relative anterior microphthalmos group had previously undergone yttrium-aluminum-garnet peripheral iridectomy. Among eyes with nanophthalmos, 2 eyes had a previous history of trabeculectomy. One of the nanophthalmic eyes required an iris retractor because of posterior synechiae. In 1 eye among patients with nanophthalmos, an Orbscan and speculometry could not be performed because of corneal edema.
|Characteristic||Nanophthalmos (n = 17)||Relative Anterior Microphthalmos (n = 29)||Normal Control (n = 54)||P Value d||Post Hoc|
|Age (years) c||55.44 ± 15.80||73.36 ± 10.06||68.96 ± 11.68||.001 a||Nanophthalmos < relative anterior microphthalmos/normal control|
|Proportion of male subjects (%)||0%||11.5%||74.1%||<.001 b|
|Right eye, n (%)||6 (35.2%)||14 (48.2%)||27 (50.0%)||.355 b|
|Mean axial length (mm) c||19.53 ± 1.17||22.19 ± 0.59||23.17 ± 0.85||<.001 a||Nanophthalmos < relative anterior microphthalmos < normal control|
|Mean ACD (mm) c||1.82 ± 0.31||1.87 ± 0.24||2.70 ± 1.31||<.001 a||Nanophthalmos/relative anterior microphthalmos < normal control|
|Mean CD (mm) c||11.16 ± 0.75||10.57 ± 0.42||11.32 ± 1.22||.005 a||Relative anterior microphthalmos < nanophthalmos/normal control|
|Mean preoperative endothelial cell count (cells/mm 2 ) c||2443.35 ± 743.02||2558.50 ± 464.38||2686.53 ± 408.80||.491 a|
|Mean preoperative IOP (mm Hg) c||18.50 ± 8.67||16.23 ± 3.8||14.62 ± 3.08||.061 a|
|Associated pathology, n (%)|
|Glaucoma||7 (41.1%)||9 (31.0%)||1 (1.8%)||<.001 b|
|Posterior synechia||2 (11.7%)||0||0||.029 b|
Table 2 shows the preoperative and the postoperative BCVA in the nanophthalmos group, the relative anterior microphthalmos group, and the normal control group. The BCVA improved postoperatively in all 3 groups, and this was statistically significant ( P = .001, P < .001, and P < .001, respectively).
|Characteristic||Nanophthalmos (n = 17)||Relative Anterior Microphthalmos (n = 29)||Normal Control (n = 54)|
|Preoperative BCVA (logMAR) a||0.67 ± 0.42||0.53 ± 0.53||0.40 ± 0.26|
|Postoperative BCVA (logMAR) a||0.41 ± 0.36||0.22 ± 0.22||0.11 ± 0.16|
|P value b||.001||<.001||<.001|
With regard to the mean numeric errors, there was a significant difference among the 3 groups based on the SRK II, with a greater myopic shift in the nanophthalmic eyes ( P = .001), and there was a greater hyperopic shift in the nanophthalmic eyes based on the Hoffer Q ( P = .024). Within the nanophthalmos group, the Holladay 1 produced the best refractive results as measured by mean numeric error ( P < .001; Table 3 ).
|IOL Formula||Mean Numeric Difference, Predicted vs Actual Postoperative SE Refraction (Diopter) ± SD|
|Nanophthalmos (n = 15)||Relative Anterior Microphthalmos (n = 29)||Normal Control (n = 54)||P Value a||Post Hoc|
|SRK II d||−1.82 ± 2.95||−0.17 ± 0.90||0.08 ± 0.48||.001 e||Nanophthalmos < relative anterior microphthalmos/normal control|
|SRK/T d||−0.28 ± 1.52||0.13 ± 0.90||0.31 ± 0.43||.152|
|Hoffer Q d||1.08 ± 1.47||0.08 ± 0.90||0.35 ± 0.47||.024 e||Nanophthalmos > relative anterior microphthalmos|
|Holladay 1 d||0.08 ± 1.62||0.09 ± 0.89||0.35 ± 0.45||.621|
|P value b||<.001 e||.001 e||<.001 e|
|Post hoc c||SRK II < SRK/T < Holladay 1 < Hoffer Q||SRKII < SRK/T/Holladay 1||SRKII < SRK/T/Hoffer Q/Holladay 1|
Nanophthalmic or relative anterior microphthalmic eyes showed the worst refractive results as measured by mean absolute error. No formula was significantly more accurate than the others as measured by mean absolute error in eyes with nanophthalmos and relative anterior microphthalmos ( Table 4 ).
|IOL Formula||Mean Absolute Difference, Predicted vs Actual Postoperative SE Refraction (Diopter) ± SD|
|Nanophthalmos (n = 15)||Relative Anterior Microphthalmos (n = 29)||Normal Control (n = 54)||P Value a||Post Hoc c|
|SRK II d||1.91 ± 2.88||0.73 ± 0.53||0.38 ± 0.29||.002 e||Nanophthalmos/relative anterior microphthalmos > normal control|
|SRK/T d||0.97 ± 1.17||0.79 ± 0.43||0.45 ± 0.29||.005 e||Relative anterior microphthalmos > normal control|
|Hoffer Q d||1.34 ± 1.23||0.74 ± 0.50||0.47 ± 0.34||.001 e||Nanophthalmos > normal control|
|Holladay 1 d||1.08 ± 1.17||0.76 ± 0.45||0.46 ± 0.34||.003 e||Relative anterior microphthalmos > normal control|
|P value b||.233||.974||.014 e|