ORIGINAL ARTICLE Year : 2014  Volume : 62  Issue : 7  Page : 788791 Personal Aconstant in relation to axial length with various intraocular lenses Mohamed A Eldaly, Khaled A Mansour Department of Ophthalmology, Faculty of Medicine, Cairo University, Giza, Egypt Correspondence Address: Purpose: To study the relationship between the axial length and personal Aconstant for the 1piece Tecnis (Abbott ZCB00), AcrySof MA60AC (Alcon) and the Quatrix aspheric preloaded (CROMA) intraocular lenses (IOL). Materials and Methods: Patients matching the inclusion criteria were further subdivided according to the implanted IOL in this prospective comparative study. The obtained refractive outcomes were introduced into the formula installed in the biometry machine (Humphrey model 820 ultrasonic biometer) to obtain the personal Aconstant for each eye. Polynomial regression analysis was done to study the individualized Aconstant for each type of IOL in relation to preoperative axial length measurement. Results: Two hundred and forty five eyes of 186 patients were enrolled into this study, of whom 73 eyes with Tecnis 1piece, 116 eyes with MA60AC, and 56 eyes with Quatrix. The median of personalized Aconstant for Tecnis 1piece, MA60AC, and Quatrix were 119.21 (SD 1.3, Std. Mean error 0.15), 119 (SD 1.2, Std. Mean error 0.11) and 120.4 (SD 1.2, Std. Mean error 0.16) respectively. Regression plots for the same range of axial length among all the groups showed that the Tecnis1 group followed the same pattern of the Quatrix group in which there was a linear relationship of a trend towards myopia when the axial length had increased and a hyperopic shift when decreased. This relationship changed into a plateau when the axial length became in the range of 23.5 mm to 27 mm in the MA60AC group. Conclusions: Personal Aconstant follows different trends with different IOLs even for the same range of axial length.
Materials and Methods Eyes of patients matching the inclusion criteria were enrolled in this prospective comparative study. Inclusion criteria were patients with dense senile cataract undergoing uncomplicated phacoemulsification through 2.8 mm clear corneal incision, and inthebag foldable intraocular lens implantation. The IOLs were either 1piece Tecnis (Abbott ZCB00), the 3piece AcrySof MA60AC (AcrySof, Alcon), or the onepiece Quatrix (QUATRIX aspheric preLoaded, CROMA) IOLs (Manufacturers' Aconstant were 118.8, 118.4 and 119.3 respectively). The IOLs were assigned using randomly generated computer numbers. All surgeries were performed by the same surgeon. All eyes have undergone preoperative biometry, performed by the same examiner, using the same Javal keratometer and acoustic biometer (Humphrey model 820 ultrasonic biometer). The refractive outcomes were obtained at least two weeks postoperatively, and the postoperative sphere and cylinder were introduced in the formula installed on the same biometry machine to obtain the personal Aconstant for each eye. The biometer utilized the SRK/T formula to calculate the IOL power preoperatively. Personal Aconstant was automatically computed by the biometer as a default accessory in model 820. The same person who performed all the surgeries was not involved in analyzing and interpreting the results to avoid performance bias. Because of the difference in the range of available powers among IOLs (+6 to +30 for Tecnis1 and MA60AC, +10 to +30 for Quatrix IOLs), lower power IOLs for eyes with higher axial lengths were not available for the Quatrix group. Subsequently, comparative analysis of IOLs was also done after adjustment for the same range of axial length among all groups. The research conforms to the tenets of the Helsinki declaration. Data were recorded on an electronic template specially designed for this study. Data were statistically represented in terms of range, mean, median, standard deviation (SD) and standard error of mean. Whenever required, comparisons were done using analysis of variance for parametric data. A probability value (P value) less than 0.05 were considered significant. For comparing nonparametric data, Chisquare and KruskalWallis tests were performed. Polynomial regression analysis was done to study the relationship of the personal Aconstant for each type of IOL to the measured preoperative axial length. Regression analysis was also done after adjusting data for the same range of axial length among studied IOLs. All statistical calculations were done using computer program›s Microsoft Excel version 7 (Microsoft Corporation, NY) and SPSS (Statistical package for the social science) and statistical programs (SPSS Inc., Chicago, IL). Results Two hundred and forty five eyes of 186 patients were enrolled into this study. They were grouped according to type of IOL used: 73 eyes (51 patients) were implanted with Tecnis 1piece (Abbott ZCB00), 116 eyes (92 patients) with 3piece AcrySof MA60AC (AcrySof, Alcon), and 56 eyes (43 patients) were implanted with onepiece Quatrix IOLs (QUATRIX aspheric PreLoaded, CROMA). The axial lengths and personal Aconstants for each type of IOL were shown in [Table 1]. Differences of axial lengths between types of used IOLs were not significant (P = 0.111). Statistically significant negative correlation existed between MA60AC personalized A constant and the corresponding axial length (0.292, P = 0.001). The same correlation was higher with the Tecnis 1piece and Quatrix IOLs (–0.311, P = 0.007; 0.548, P < 0.0001 respectively). The fit of the curve for polynomial regression analysis was highly significant with the three groups of IOLs (P < 0.0001). Plots of regression for personalized Aconstant with the axial length in Tecnis 1piece, MA60AC and Quatrix IOLs were illustrated in [Figure 1], [Figure 2] and [Figure 3] respectively. Inspection of the data in the figures illustrated that this negative correlation did not extend across the entire range of axial length. It appears that the relation of the Aconstant to axial length is more complex and that the Aconstant became positively correlated with axial lengths above 27 mm. There was an interval between 24.527.5 mm outside which there was a hyperopic shift in the Tecnis 1piece group [Figure 1]. The same pattern was detected in the MA60AC group with a lesser steeper curve compared to Tecnis 1, outside the range 24.528.5 mm [Figure 2]. The plots for the Quatrix group showed a different pattern with a linear relationship of a trend towards myopia when the axial length had increased and a hyperopic shift when decreased [Figure 3].{Figure 1}{Figure 2}{Figure 3}{Table 1}{Table 2} Comparative analysis was done after adjusting the data for the same range of axial length among all groups, 21.527 mm [Table 2]. The fit of the curve for polynomial regression analysis was significant with the three groups of IOL (P < 0.0001, P = 0.002 and P < 0.0001 for the Tecnis 1, MA60AC and Quatrix groups respectively). Plots of regression for personalized Aconstant with the adjusted range of axial length in Tecnis 1piece, MA60AC and Quatrix IOLs were illustrated in [Figure 4], [Figure 5], [Figure 6] respectively. The plots for Tecnis 1 group followed the same pattern of the Quatrix group plots in which there was a linear relationship of a trend towards myopia when the axial length had increased and a hyperopic shift when decreased [Figure 4] [Figure 6]. On the other hand, In the MA60AC group, this relationship changed into a plateau when the axial length became in the range of 23.5 mm to 27 mm [Figure 5].{Figure 4}{Figure 5}{Figure 6} Discussion Various reports advised the continuous review of refractive outcomes as part of the quality control management. They added that optimization of Aconstant helps improving the refractive outcomes. Few studies were published regarding the optimization and even fewer ones that have studied the trends of personalization of manufacturers' Aconstant in different types of IOLs. [6],[7],[8] Langenbucher and his colleagues recommended standardizing the practice setting and that all cases to be operated by a single surgeon for valid individualization of Aconstant. [6] On the other hand, Aristodemous et al. described that the differences in personalized Aconstant between most surgeons as clinically insignificant. They further added that optimization for an IOL was far exceeding any additional benefit for individual surgeon personalization. [9] A recent report described that different types of IOL show different trends even with standardization of the same practice settings. [6] Haigis reported the strong dependence of Aconstants on axial length. He emphasized that even for the same IOL and same instrumentation, different IOL constants were necessary for different ranges of axial lengths. [10] These findings were in agreement with those of our study. A hyperopic shift was reported in the surgical experience with Tecnis 1piece IOL. That was explained due to the special design of the IOL haptics that allowed a potential for the optic to move posteriorly. That was not reported with AcrySof IOL (IQ) due to a more appropriate design of lens haptics. [11] In our study, we found a trend towards hyperopia outside a range of axial lengths when personalizing the Aconstant for the Tecnis 1piece and AcrySof MA60AC IOLs. However, there were a small number of patients with higher axial lengths as the lowest possible IOL powers for Tecnis 1 and MA60AC was + 6, while for Quatrix was + 10. Subsequently, it was not possible to study the trend of personal Aconstant with the Quatrix IOL for axial lengths beyond 27 mm. In the comparative analysis using the same range of axial lengths between the three groups, Tecnis 1 and Quatrix groups followed the same trend. With the earlier versions of SRK formula, a tendency towards hyperopia had been reported with axial lengths shorter than 22.0 mm, and myopia when longer than 24.5 mm. [12] The SRK/T formula had been developed to overcome this error. Eom et al. reported the same trends which we found with Tecnis 1 and Quatrix IOLs, when they studied AcrySof IQ (similarly, single piece). They attributed this finding to the differences they found in corneal curvature among their patients. [13] Nejima et al. reported that MA60AC (3piece) showed some forward shift when compared to singlepiece IOLs. [14] This forward shift could possibly induce some myopia that counteracted with the comparative hyperopia found with the same range of axial lengths in Tecnis 1 and Quatrix groups. This could provide some explanation to the plateau found in this study with MA60AC. In conclusion, different IOLs showed different trends when personalizing their Aconstant even for the same range of axial lengths. Analyzing the same comparative range of axial length among all groups, MA60AC showed no significant myopic/hyperopic shift within the range of 23.527 mm. A linear trend was found towards myopia with the increase of axial length in the Tecnis 1 and Quatrix groups. Further studies are needed to investigate of these relationships with various IOLs. References


