Astigmatism with long AL is the common refractive error of patients with cataract. Toshiyuki Miyake et al. have studied the distribution of corneal astigmatism in 12,428 eyes after cataract surgery and reported that 36.3% of eyes have more than 1.0 D of corneal astigmatism, 8.0% have more than 2.0 D, and 2.4% have more than 3.0 D . Postoperative misalignment is a significant problem in toric IOL implantation. Therefore, we observed the clinical outcomes and evaluated rotational stability and its possible related factors, including preoperative biometric examinations and surgical operation.
In our retrospective study, the mean RAS and absolute IOL rotation were 0.91±0.74 D, ranging from 0 to-4.5D, and 7.48±11.19° (0-79°) at 3 months after surgery, respectively, which were larger than those in the eyes with various axial length in other studies[5,11]. This might be attributed to the fact that all patients had long AL or a higher percentage of high myopia. No significant differences were observed in toric IOL rotation between 1 and 3 months, suggesting that toric IOL rotation mainly occurred within 1 month after surgery . Compared with toric IOL implantation in eyes of emmetropia, hyperopia or AL shorter than 25 mm of other studies [13-16], the higher rate of IOL rotation would occur in patients with long AL probably.
We established that Toric IOL rotation was positively correlated with the area of capsulorhexis (y=1.008x-13.724, Pearson’s r=0.297, p=0.017), while there was no correlation between the overlapped area between IOL optic and capsulorhexis. These findings suggested that an appropriately sized CCC was essential to prevent IOL rotation, especially at early stage after surgery. A relatively smaller sized CCC was conducive than oversize one. There were significant negative correlations between the area of CCC and the overlapped area between IOL optic and capsulorhexis (p<0.001, r=-0.979). Moreover, we didn't find a correlation between IOL rotation and AL, LT, VL, ACD, STS, keratometry value, the spherical power of the implanted IOL.
Four cases in our study rotated more than 15°. There were no significant differences between the four cases and other cases in terms of average ACD, ACV, LT, AL, the intraoperative area of CCC, and the area of overlap between IOL optic and capsulorhexis. Among the data, the case in which toric IOL rotation reached 79° at 3 months after surgery had a significantly larger intraoperative area of capsulorhexis than the average of others. The capsulorhexis size was too large to fix the toric IOL, and the edge of IOL was almost free, therefore leading to IOL rotated extensively (Fig.3). It was observed in the two cases which rotated 33° and 40° without too large capsulorhexis size, including larger preoperative corneal astigmatism, intraoperative approximately vertical toric IOL alignment. No correlations were found between IOL rotation and the variables, including keratometric value, intraoperative IOL axial direction, preoperative corneal astigmatism. Therefore, accumulation of more cases and a long-term observation should be taken in future.
During the first 3 months after surgery, several studies have mentioned that the toric IOL rotation can be attributed to an IOL surface with a low coefficient of friction , thinner optic of low-powered IOLs , the design of the IOL , an overall length of the IOL that is too small for the capsular bag  and instability of the anterior chamber .
Some earlier studies have found a positive correlation between early rotation of the toric IOL and AL, which is positively correlated with the diameter of the capsular bag [21-23]. Since the diameter of capsular bag cannot be measured directly, we measured STS using Anterior Segment VisanteTM OCT, which is a non-contact optical signal acquisition. We didn't find a correlation between AL and STS (Pearson’s r=-0.005, p=0.971), suggesting that AL was not positively correlated with the diameter of the capsular bag in a long AL eye.
Other earlier literature has also indicated that the area of CCC was contractible at 3 months compared with 1 month after surgery. When the collapses and fibroses of capsular bag occur, the area of CCC would decrease and influenced the IOL rotation stability . In our observation, compared with intraoperative one, the area of CCC was decreased slightly at 1 month postoperatively and continued to be decreased at 3 months postoperatively. Meanwhile, the axis of implanted toric IOL was not changed in the 3-month follow-up.
There are some limitations in our study. The digital anterior segment photographs were used to calculate the area of capsulorhexis as well as the overlapped area between capsulorhexis and optic. This process was related with the subjectivity of the operator. In order to reduce the influence of inaccurate calculation, we repeated the measurement for three times. The toric IOL alignment axis marker was at 10° intervals, and manual marking was used. VERION system has the advantage of intraoperative digital guidance of the toric IOL alignment even for 1° . It has been reported that the underlying capsular shrinking probably occurs at 6 months or 1 year after cataract surgery, with the beginning in the first 3 months. In order to reduce the postoperative RAS seized with SIA, we evaluated the early-stage outcomes until 3 months after the surgery in this study. Inevitably, some cases might already suffer from a very early-stage fibrosis and contraction of the capsular bag at this time.
Our data showed the refractive results and the factors influencing toric IOL rotation in eyes with long AL. The area of CCC affected their early-stage stability of toric IOL axis after surgery. However, the long-term influence from the capsular bag contraction and posterior capsular opacity should be considered.