Demographics
In the present study, 60 eyes of 56 patients with 1.8-mm steep-axis CCI, including 24 males and 32 females, were enrolled. The age of patients ranged from 43 to 82 years with an average of 58.8 ± 10.7 years. The follow-up period ranged from 4.0 to 13.6 weeks with an average of 8.7 ± 3.4 weeks.
Anterior corneal astigmatism
The numbers of eyes with WTR astigmatism, ATR astigmatism and oblique astigmatism of the preoperative anterior corneal were 34, 24 and 2, respectively. The number of eyes with the preoperative anterior corneal astigmatism less than 1.5 D accounted for 48.3 %, that of ≥1.5 D and < 2.5 D accounted for 43.3%, and that of more than 2.5 D accounted for 8.3% (Table 1).
The preoperative anterior corneal astigmatism ranged from 0.70 D to 3.70 D with an average of 1.58±0.61 D. The mean preoperative anterior corneal K1 was 43.36±1.67 D, and K2 was 44.96±1.64 D. The postoperative anterior corneal astigmatism ranged from 0.30 D to 3.30 D, with an average of 1.26±0.68 D. The postoperative K1 was 43.32±1.74 D, and K2 was 44.61±1.68 D. Paired t-test showed a significant reduction in postoperative anterior corneal K2 compared with the preoperative one (P< 0.001), and no significant difference was found in postoperative anterior corneal K1 and the preoperative one (P= 0.395). Paired t-test showed a significant reduction in postoperative anterior corneal astigmatism compared with the preoperative one (P< 0.001) (Table 2). Fig. 1 shows the preoperative and postoperative astigmatism distribution of anterior cornea. A concentration trend could be found between Fig. 1A and Fig. 1B.
Posterior corneal astigmatism
The numbers of eyes with WTR astigmatism, ATR astigmatism and oblique astigmatism of the preoperative posterior corneal were 34, 10 and 9, respectively, of which WTR astigmatism accounted for 56.7%. There were 49 eyes with a preoperative posterior corneal astigmatism less than 0.5 D, accounting for 81.7% (including seven eyes with an astigmatism of 0 D), and there were 11 eyes with an astigmatism of more than 0.5 D and less than 1.0 D, accounting for 18.3% (Table 1).
The mean preoperative posterior corneal K1 was -6.28 ± 0.25 D, and K2 was -6.55 ± 0.27 D. The posterior corneal astigmatism ranged from 0.00 D to 0.80 D with an average of 0.28 ± 0.22 D. The postoperative posterior corneal astigmatism ranged from 0.00 D to 1.10 D with an average of 0.41±0.26 D. The postoperative K1 was -6.33 ± 0.28 D, and K2 was -6.75 ± 0.37 D. Paired t-test showed a significant increase in postoperative posterior corneal K1 (P= 0.003), K2 (P< 0.001), and astigmatism (P< 0.001) compared with the preoperative one (Table 2). Fig. 2 shows the preoperative and postoperative astigmatism distribution of posterior cornea. A significant decentration pattern was noted between Fig. 2A and Fig. 2B.
Total corneal astigmatism
The numbers of eyes with WTR astigmatism, ATR astigmatism and oblique astigmatism of the preoperative total astigmatism corneal were 34, 23 and 3, respectively. The number of eyes with the anterior corneal astigmatism less than 1.5 D accounted for 35.0 %, that of ≥1.5 D and < 2.5 D accounted for 56.7%, and that of more than 2.5 D accounted for 8.3% (Table 1).
The preoperative total corneal astigmatism ranged from 1.00 D to 3.80 D with an average of 1.70 ± 0.52 D. The postoperative total corneal astigmatism ranged from 0.75 D to 3.40 D with an average of 1.30 ± 0.51 D. Fig. 3 shows the preoperative and postoperative total astigmatism distribution. A significant decentration pattern was noted between Fig. 3A and Fig. 3B.
P-SIA and its correlation analysis
P-SIA ranged from 0.00 D to 0.70 D with an average of 0.34 ± 0.20 D. There were 44 eyes with a P-SIA of less than 0.5D, accounting for 73.3%, while the P-SIA of 26.7% eyes was not less than 0.5 D (Table 4). Fig. 4 reveals that P-SIA had no concentration trend in a certain direction on the polar plot.
According to the astigmatic axis of the posterior cornea, the corneal astigmatism was divided into three groups: WTR group (n=34), ATR group (n=10) and the oblique astigmatism group (n=9). The P-SIA of the WTR group, ATR group and oblique astigmatism group was 0.36±0.21 D, 0.38±0.17 D and 0.31±0.18 D, respectively. There was no significant difference among the groups by using one-way ANOVA (F=0.335, P=0.717).
The mean P-SIA of male group and female group was 0.34±0.19 D and 0.34±0.20 D, respectively. There was no significant difference between the groups by using Independent t-test. (P =0.623)
According to the follow-up time, the corneal astigmatism was divided into three groups: 4 ~ 8w group (n=15), 8 ~12w group (n=25) and more than 12w group (n=20). The P-SIA of 4 ~ 8w group, 8 ~12w group and more than 12w group was 0.39 ± 0.19 D, 0.35 ± 0.18 D and 0.30 ± 0.23 D respectively. There was no significant difference among the groups by using one-way ANOVA (F=0.706, P =0.498).
Pearson correlation analysis was performed between P-SIA and anterior/posterior corneal astigmatism, K1 and K2, and the following results were obtained. (1) P-SIA was significantly and positively correlated with anterior corneal astigmatism (r = 0.29, P =0.024) (Fig. 5). (2) There was no significant correlation between P-SIA and anterior corneal K1 (r = 0.07, P =0.576). (3) There was no significant correlation between P-SIA and anterior corneal K2 (r=0.07, P =0.608). (4) P-SIA was significantly and positively correlated with the posterior corneal astigmatism (r=0.27, P =0.038) (Fig. 6). (5) There was no significant correlation between P-SIA and posterior corneal K1 (r=0.04, P =0.789). (6) P-SIA was not significantly correlated with posterior corneal K2 (r = 0.18, P = 0.163).
In order to investigate whether the P-SIA was smaller in patients with closer axis of anterior and posterior corneal astigmatism, we performed the correlation analysis between the P-SIA and absolute axis differences of the corneal anterior and posterior astigmatism. There was no significant correlation between P-SIA and the absolute difference (r=0.03, P =0.808).
Multivariable regression analysis of P-SIA
Age, sex, follow-up time, preoperative anterior corneal K1, K2 and astigmatism, preoperative posterior corneal K1, K2 and astigmatism, and the absolute axis differences of the corneal anterior and posterior astigmatism were incorporated as covariates in the multivariable regression model to adjust for their effects on P-SIA. The final multivariable model, which included preoperative anterior corneal astigmatism (B=0.347, P=0.005) and preoperative posterior corneal astigmatism (B=0.298, P=0.014), had the best-fit index with R2=0.205. In this multivariable model, the preoperative anterior and posterior corneal astigmatism showed a significant effect on P-SIA (F=7.344, P=0.001).