Introduction

Keratoconus is a progressive disorder characterized by ectasia and thinning of the cornea. The progressive thinning and subsequent anterior protrusion of the cornea can result not only in severe myopic astigmatism but also asymmetrical irregular astigmatism, leading to distorted vision. Considering that keratoconic eyes often show some amount of astigmatism and tend to develop cataracts earlier than non-keratoconus eyes [1], toric intraocular lens (IOL) implantation appears to be a good surgical approach for the correction of spherical and cylindrical errors for cataractous eyes with keratoconus. There are several studies of toric IOL implantation for mild keratoconus [2–7], but most studies have merely focused on the astigmatic outcomes of toric IOL implantation for these eyes. Since this surgery does not aim to reduce higher-order aberrations (HOAs), the number of corneal HOAs is likely to play a vital role in the postoperative visual performance of such keratoconic patients. However, so far the analysis of corneal HOAs has not been performed in such eyes. The purpose of the current study was to prospectively assess the clinical outcomes, including the changes in astigmatism and corneal HOAs, of phacoemulsification with toric IOL implantation in cataractous eyes with mild non-progressive keratoconus.

Methods

Study population

This prospective study comprised 19 eyes of 19 cataractous and keratoconic patients (6 men and 13 women, mean age ± standard deviation, 63.1 ± 9.1 years), with corneal astigmatism of 1.25 diopters (D) or more, who underwent successful phacoemulsification with toric IOL implantation (AcrySof IQ Toric, Alcon Laboratories Inc, Ft Worth, TX, USA), with good quality scans of corneal topography measured with the Atlas corneal topographer (Carl Zeiss Meditec, Oberkochen, Germany). The patients were recruited in a continuous cohort. One eye per subject was selected randomly for statistical analysis. Keratoconus was diagnosed by one experienced clinician (K.K.) based on evident findings characteristic of keratoconus (e.g., corneal topography with asymmetric bow-tie pattern with or without skewed axes), and at least one keratoconus sign (e.g., stromal thinning, conical protrusion of the cornea at the apex, Fleischer ring, Vogt striae, or anterior stromal scar) on slit-lamp examination [8]. All patients had rigid gas-permeable lens intolerance. We confirmed stable keratometry and refraction (<0.5 D change) for at least 6 months. All eyes had grade I or II keratoconus according to the Amsler-Krumeich classification, based on astigmatism, corneal power, corneal transparency, and corneal thickness [9]. We determined uncorrected distance visual acuity (UDVA), corrected distance visual acuity (CDVA), refractive astigmatism, corneal astigmatism, corneal HOAs, and astigmatic axis rotation both preoperatively and 3 months postoperatively. Corneal astigmatism was measured with the Atlas and corneal HOAs with the KR9000-PW (Topcon, Tokyo, Japan). The study was approved by the Institutional Review Board of Kitasato University and followed the tenets of the Declaration of Helsinki. Informed consent was obtained from all patients after explanation of the nature and possible consequences of the study.

Intraocular lens selection

A web-based toric IOL calculator program was used to determine the optimal cylinder power and alignment axis of the IOL (http://www.acrysoftoriccalculator.com). IOL power calculations were performed by the SRK/T formula using the axial length measured by a partial coherence interferometer (IOL Master, Carl Zeiss Meditec, Jena, Germany) and the keratometric readings on the 3.0-mm ring measured by the corneal topographer (Atlas), without any correction.

Surgical procedure

We used the axis registration technique to correct ocular cyclotorsion, as described previously [10]. In brief, preoperatively, a limbal reference mark was placed using an axis marker (AE-2748; ASICO, Westmont, IL, USA) under a slit-lamp with the patient sitting upright. The marking on the conjunctiva was easily identified on the topographic image (TMS-5, Tomey, Aichi, Japan), allowing us to confirm the graphic relation between the mark and the steepest meridian. Intraoperatively, the implantation axis was determined using the reference mark and the alignment axis obtained from the manufacturer’s program, and was marked using a Mendez ring and a 30-gauge needle. A standard phacoemulsification was performed by capsulorhexis, nuclear and cortex extraction; a toric IOL (SN6AT, Alcon Laboratories Inc., Fort Worth, TX, USA) was implanted in the capsular bag through a 2.65-mm temporal corneal incision [11]. All surgery was uneventfully performed by two experienced surgeons (K.S., K.K.) using the same technique. In 6 of 19 eyes, near emmetropia was selected as the target refraction to reduce the preoperative refractive errors. In the remaining 13 eyes, undercorrection for near vision was intentionally selected. No adjustments to the manufacturer’s nomograms were done. Postoperatively, steroidal (0.1 % betamethasone, Rinderon™, Shionogi, Osaka, Japan), antibiotic (levofloxacin, Cravit™, Santen, Osaka, Japan), diclofenac sodium (0.1 % Diclod™, Wakamoto, Tokyo, Japan) medications were topically administered 4 times daily for 1 month, and then the dose was steadily reduced.

Digital anterior segment photographs of sufficient quality were taken immediately postoperatively and 3 months postoperatively, as described previously [11]. Either a blood vessel of the bulbar conjunctiva, pigment of bulbar conjunctiva, or iris pattern of the photographs was selected as a reference point, and the amount of postoperative IOL rotation was determined by comparing immediate and 3-month postoperative photographs using this reference point.

Power vector analysis

Spherocylindrical refraction results were converted to vectors expressed by 3 dioptric powers: M, J 0, and J 45, where M is equal to the spherical equivalent of the given refractive error and J 0 and J 45 are the 2 Jackson cross-cylinder equivalents to the conventional cylinder. Manifest refractions were recorded in conventional script notation (sphere, cylinder, and axis) and then converted to the power vector coordinates described by Thibos and Horner [12] and to overall blurring strength by the following formulas:

$$ M \, = \, S + C/2, $$
$$ J_{0} = \, \left( { - C/2} \right)\cos \left( {2\alpha } \right), \, J_{45} = \, \left( { - C/2} \right)\sin \left( {2\alpha } \right), $$
$$ B \, = \, \left( {M^{2} + J_{0}^{2} + J_{45}^{2} } \right)^{1/2} , $$

where M is the spherical lens equal to the spherical equivalent of the given refractive error; S is the sphere; C is the cylinder; J 0 is the Jackson cross-cylinder, axes at 180° and 90°; α is the axis; J 45 is the Jackson cross-cylinder, axes at 45° and 135°; and B is the overall blurring strength of the spherocylindrical refractive error.

Statistical analysis

All statistical analyses were performed using commercially available statistical software (Ekuseru-Toukei 2010, Social Survey Research Information Co, Ltd., Tokyo, Japan). The normality of all data samples was first checked by the Kolmogorov–Smirnov test. Since the data did not fulfill the criteria for normal distribution, the Wilcoxon signed-rank test was used for statistical analysis to compare the pre- and post-surgical data, and the Spearman correlation coefficient was used to check the relationship between the two variables. Unless otherwise indicated, the results are expressed as mean ± standard deviation, and a value of p < 0.05 was considered statistically significant.

Results

Study population

Preoperative demographics of the study population are summarized in Table 1. All surgery was uneventful, and no definite intraoperative complication was observed. No eyes were lost during the 3-month follow-up in this series.

Table 1 Preoperative demographics of the study population in eyes undergoing toric intraocular lens implantation for mild keratoconus
Full size table

Visual and refractive outcomes

Visual and refractive outcomes of toric IOL implantation in each patient are listed in Table 2. Logarithm of the minimal angle of resolution (logMAR) UDVA was significantly improved from 1.14 ± 0.50 (range, 0.15–2.00) preoperatively to 0.46 ± 0.33 (range, −0.08 to 1.15) 3 months postoperatively (Wilcoxon signed-rank test, p < 0.001). In 5 (83 %) of 6 eyes for which the target refraction was emmetropia, the postoperative UDVA was better than 20/32. LogMAR CDVA was also significantly improved from 0.27 ± 0.45 (range, −0.08 to 2.00) preoperatively to −0.01 ± 0.09 (range, −0.18 to 0.15) postoperatively (p < 0.001). The refractive astigmatism was significantly decreased from −1.92 ± 1.73 D (range, 0.00 to −5.00) preoperatively to −0.70 ± 0.60 D (range, 0.00 to −2.00) postoperatively (p = 0.006).

Table 2 Visual and refractive outcomes of toric intraocular lens implantation in eyes with mild keratoconus
Full size table

Astigmatism and higher-order aberrations

The corneal astigmatism was not significantly changed from 2.89 ± 1.30 D (range, 1.37–6.12) preoperatively to 2.98 ± 1.09 D (range, 1.75–5.37 D) postoperatively (p = 0.492). The corneal HOAs for a 4-mm pupil were not significantly changed from 0.47 ± 0.23 µm (range, 0.17–0.99) preoperatively to 0.52 ± 0.26 µm (range, 0.09–1.11) postoperatively (p = 0.211). We found no significant correlation between logMAR CDVA and corneal HOAs preoperatively (Spearman correlation coefficient r = 0.107, p = 0.662), but a significant correlation between them 3 months postoperatively (r = 0.514, p = 0.024).

Power vector analysis

The changes in the astigmatic power vector between preoperative and postoperative values of corneal and refractive astigmatism for all cases are presented in Figs. 1 and 2, respectively. For corneal astigmatism, the dispersed cluster of points before surgery tended to remain unchanged after surgery. For J 0, 32 % of eyes were within ±0.5 D, and 58 % were within ±1.0 D. For J 45, 11 % of eyes were within ±0.5 D, and 58 % were within ±1.0 D postoperatively. For refractive astigmatism, the dispersed cluster of points before surgery tended to collapse around the origin after surgery. For J 0, 95 % of eyes were within ±0.5 D, and 100 % were within ±1.0 D. For J 45, 89 % of eyes were within ±0.5 D, and 100 % were within ±1.0 D postoperatively. A scatter plot of the attempted versus the achieved astigmatic correction 3 months postoperatively is shown in Fig. 3. We found a significant correlation between corneal HOAs and the astigmatic correction (r = 0.503, p = 0.028 for J 0, and r = 0.465, p = 0.045 for J 45).

Fig. 1
figure 1

Power vector analysis of corneal astigmatism preoperatively and 3 months postoperatively, plotted as an astigmatic vector for each eye, referenced to the spectacle plane, in eyes with cataract and keratoconus

Full size image
Fig. 2
figure 2

Power vector analysis of refractive astigmatism preoperatively and 3 months postoperatively, plotted as an astigmatic vector for each eye, referenced to the spectacle plane, in eyes with cataract and keratoconus

Full size image
Fig. 3
figure 3

A scatter plot of the attempted versus the achieved astigmatic correction 3 months postoperatively in eyes with cataract and keratoconus

Full size image

Predictability

A scatter plot of the attempted versus the achieved refraction (manifest spherical equivalent) 3 months postoperatively is shown in Fig. 4. Thirteen (68 %) eyes were within ±0.5 D and 18 eyes within 1.0 D, of the targeted correction.

Fig. 4
figure 4

A scatter plot of the attempted versus the achieved refraction (manifest spherical equivalent) 3 months postoperatively in eyes with cataract and keratoconus. Thirteen (68 %) and 18 (95 %) of 19 eyes were within ±0.5 and 1.0 D, respectively, of the targeted correction

Full size image

Intraocular lens rotation

The mean amount of the postoperative IOL rotation was 5.2 ± 2.6° (range, 1°–10°). No eyes in this series needed IOL repositioning.

Secondary surgeries/adverse events

Only one eye (5 %) developed posterior capsular opacification, requiring an Nd-YAG capsulotomy. Neither cataract incision-related complications, keratectasia, nor other vision-threatening complications were seen at any time during the 3-month observation period.

Discussion

Our results demonstrate that toric IOL implantation was effective for the correction of astigmatic errors in cataractous eyes with mild non-progressive keratoconus without a significant induction of corneal HOAs. Although cases 1 and 2 had higher amounts of corneal astigmatism (4.38 D for case 1, and 3.87 D for case 2), we used the T5 toric IOL model for astigmatic correction, since the T6–T9 models were not available at that time in Japan. Therefore, the relatively large refractive astigmatism in these eyes was retained after toric IOL implantation (1.5 D for case 1, and 2.0 D for case 2). Although the toric IOL manufacturer recommends that careful preoperative evaluation should be performed in eyes having irregular astigmatism, our astigmatic outcomes in this study were comparable with their astigmatic outcomes in previous studies [2–7], suggesting that toric IOL implantation is a feasible surgical option for astigmatic correction in eyes with mild keratoconus. Navas et al. presented 2 cases of successful toric IOL implantation for forme fruste keratoconus and keratoconus, demonstrating that the postoperative UDVA was 20/25 in both cases, with refractions of −0.25, −0.50 × 140 and 0.25, −0.50 × 60 [2]. Visser et al. also presented 2 cases of successful toric IOL implantation for keratoconus, demonstrating that refractive astigmatism decreased by 70 % in both eyes [3]. Jaimes et al. reported, in a study of 19 eyes of 13 patients undergoing refractive lens exchange, that the refractive astigmatism was significantly decreased from 3.95 ± 1.30 D preoperatively to 1.36 ± 1.17 D postoperatively, and that logMAR UDVA was significantly improved from 1.35 ± 0.36 preoperatively to 0.29 ± 0.23 postoperatively [4]. In a study of 12 eyes of 9 mild to moderate keratoconic patients, Nanavaty et al. demonstrated that the refractive astigmatism was significantly decreased from 3.00 ± 1.00 D preoperatively to 0.70 ± 0.80 D postoperatively, and that the postoperative UDVA was 20/40 or better in 75 % and the CDVA was 20/40 or better in 83.3 % of eyes [5]. In a study of 17 eyes of 10 keratoconic patients undergoing micro-incision cataract surgery, Alió et al. showed that the refractive astigmatism was significantly decreased from 2.95 ± 1.71 D preoperatively to 1.40 ± 1.13 D postoperatively, and that logMAR UDVA was significantly improved from 1.33 ± 0.95 preoperatively to 0.32 ± 0.38 postoperatively [6]. In a study of 23 of 17 keratoconic patients, Hashemi et al. demonstrated that UCVA increased significantly in mild and moderate keratoconus groups postoperatively, and that the difference in residual astigmatism between preoperative and postoperative manifest refraction was 0.45 ± 0.36, 0.54 ± 0.39, and 2.60 ± 0.95 D in the mild, moderate, and severe keratoconus groups, respectively [7].

Since keratoconic eyes often show some amount of irregular astigmatism, which cannot be corrected by toric IOL implantation, the presence of irregular astigmatism in such eyes may lead to a deterioration in postoperative visual performance. We found a significant association between logMAR CDVA and corneal HOAs postoperatively, suggesting that corneal HOAs also play a role in visual performance in keratoconic patients undergoing toric IOL implantation. When, in the current study, we analyzed 6 eyes in which the target refraction was emmetropia targeting the amount of corneal HOAs, 5 eyes having postoperative UDVA of 20/32 or better had preoperative corneal HOAs of 0.249 to 0.353 µm, and the remaining 1 eye having postoperative UDVA of 20/100 had corneal HOAs of 0.990 µm. As far as we can ascertain, this is the first study that investigates the clinical relevance of corneal HOAs in such patients. Based on our clinical findings, the selection criteria for toric IOL implantation for keratoconus may include rigid gas-permeable lens intolerance, stable keratometry and refraction (for at least 6 months), grade I keratoconus and lower corneal HOAs. We believe that this information will be helpful in determining the surgical indication for toric IOL implantation for keratoconus in daily practice.

This study has several limitations. First, the sample data were kept rather limited and the follow-up period was set at 3 postoperative months. However, the actual number of patients undergoing toric IOL implantation for keratoconus is not very large and, based on the fact that the IOL rotation usually occurs in the early postoperative period, it is unlikely that the results changed substantially in the late postoperative period. Kim et al. reported that the mean rotation of the IOL was 3.45 ± 3.39° after a longer follow-up of more than 1 year, and that it was not significantly different from early follow-up results, suggesting that the AcrySof toric IOL has good rotational stability during the early and late follow-up periods [13]. We also demonstrated that both manifest refraction and keratometry remained unchanged 3 years after posterior chamber phakic IOL implantation for mild keratoconus, indicating that a 3.0-mm temporal corneal incision does not induce a significant change in the corneal shape, and that no progression of the disease took place even in the late postoperative period [14]. Secondly, we used only the 3-mm ring data of corneal topography-derived keratometry for toric IOL calculation in keratoconic eyes, as described previously [2–7]. Hashemi et al. demonstrated that the lowest mean absolute error was seen in the mild and moderate keratoconus groups with corneal topography-derived keratometry using the SRK/T formula [7]. However, it is sometimes difficult to accurately determine the axis orientation in keratoconic eyes having skewed hemi-meridians. Thirdly, we did not measure corneal astigmatism or corneal HOAs of the posterior surface in this study. We recently showed that the mean magnitude of anterior central corneal astigmatism was approximately 4 D and of posterior, 1 D, suggesting that the presence of posterior corneal astigmatism is not necessarily negligible for the accurate astigmatic correction in keratoconus [15]. Although the change in the posterior corneal surface plays a more subtle role in optical performance than that in the anterior corneal surface, because of the smaller change in the refractive index, the adoption of total corneal astigmatism instead of anterior corneal astigmatism may be helpful for selecting a more appropriate toric IOL model to yield more precise astigmatism correction, especially for keratoconus.

In summary, our results support the view that toric IOL implantation was effective in reducing refractive astigmatism without a significant change in corneal HOAs for mild non-progressive keratoconus with cataract, suggesting its viability as a surgical option for these patients requiring cataract surgery. A meticulous preoperative patient selection, such as the assessment of corneal HOAs, may play a key role in successful toric IOL implantation for keratoconic patients. A further study with a larger number of patients is necessary to confirm these preliminary findings.