Postoperative Rotation and Refractive Efficacy of Toric Intraocular Lenses After Phacoemulsification Cataract Surgery

Abstract

Introduction

We evaluated the rotational stability and factors associated with it in patients implanted with Eyecryl™ roric IOL after phacoemulsification surgery. We also assessed the factors associated with refractive efficacy in these patients.

Methods

The present study is a secondary analysis to study the rotation in 121 eyes from 94 patients. We included the following data: demographics; preoperative refraction and corrected distance visual acuity (CDVA), type of astigmatism, axial length, anterior chamber depth, and aqueous depth; power of the implanted IOL, axis of the implanted IOL, and predicted residual error; magnitude and direction of postoperative rotation of the IOL; postoperative vision and refraction (day 30 ± 3 days).

Results

About 86% of the eyes had postoperative astigmatism of ≤ 0.25 D and 9% had 0.26 to 0.50 D on day 30. The median (IQR) rotation of the IOL on day 30 (± 3 days) was 1 (1, 3) degree and the mean (SD) rotation was 2.31 (2.97) degrees. Eyes with high rotation had significantly higher median (IQR) preoperative sphere (more positive side) (1.50 D [0, 1.75] vs 0 D [− 1.0, 0.75]; p = 0.013) and lower mean (SD) axial length (22.6 [0.9] vs 23.2 [0.9]; p = 0.02). The mean (SD) postoperative refractive efficacy was 85.1 (13.6) %; it was significantly lower in those who had with-the-rule astigmatism compared with against-the-rule astigmatism (69.6% vs 86.7%; p = 0.046). The proportion of eyes with difference in lines between postoperative UDVA and CDVA was significantly higher in those with low postoperative efficacy (70.0% vs 21.8%; p < 0.001).

Conclusion

The Eyecryl™ toric intraocular lens had low rotation at 1-month post-surgery. Higher rotation was associated with a higher preoperative sphere (more positive side) and shorter eyes. Postoperative efficacy was higher in eyes with against-the-rule astigmatism. Eyes with < 75% efficacy and/or ≥ 5° rotation showed over-correction on vector analysis.

Key Summary Points

Why carry out this study?

The main purpose of toric intraocular lenses (IOLs) is to correct astigmatism in patients with cataract surgery; therefore, appropriate alignment and placement are important for postoperative outcomes

The effectiveness of the IOLs may be compromised if the lens rotates from its intended placement position, and rotation may be associated with surgical factors, IOL characteristics, and biological factors

What is the rotational stability and postoperative efficacy of Eyecryl™ toric IOL in patients after phacoemulsification cataract surgery?

What has been learned from the study?

The mean (SD) rotation was 2.31 (2.97) degrees, and the mean (SD) postoperative refractive efficacy was 85.1 (13.6)%

Eyecryl™ toric intraocular lens had low rotation even after 1 month post-surgery; higher rotation was associated with higher preoperative spheres (more positive side) and shorter eyes

Postoperative efficacy was higher in most eyes; this was associated with against-the-rule astigmatism

Eyes that had either lower efficacy/high rotation or both showed over-correction on vector analysis

Introduction

Toric intraocular lenses (IOLs), introduced initially in 1992, have been a useful addition to various types of lenses available to patients undergoing cataract surgery [1]. Compared with standard non-toric IOLs, toric IOLs have different optical powers in different meridians; this allows them to correct the asymmetric refractive error characteristic of astigmatism [2–4]. Use of these IOLs requires precise measurements to determine the appropriate power and orientation for the toric lens prior to surgery, thus ensuring that the lens is correctly aligned during implantation to achieve optimal visual outcomes [2, 4]. Although the indications for using toric IOLs have increased over the past few years, the common indications are age-related cataract with corneal astigmatism, especially those with ≥ 1 diopter (D), and mild to moderate regular astigmatism [2, 5]. The contraindications for using these types of IOLs are patients with any developmental abnormality or trauma that may lead to problems in capsular bag support [5].

The main purpose of toric IOLs is to correct astigmatism in patients with cataract surgery; therefore, appropriate alignment and placement is important for postoperative outcomes. The effectiveness of the IOLs may be compromised if the lens rotates from its intended placement position. Most of the rotation occurs in the first 24 h of surgery, and significant changes are noted with the first week of surgery [6]. Furthermore, in most of these patients, stability is achieved at 1-month post-surgery, and only minimal rotation is observed after this period [6, 7]. Rotation > 10° reduces the efficacy of attempted cylindrical power by 30%; thus, there are greater chances of repositioning in these patients [8]. Rotation of the toric IOLs may depend on multiple factors—such as surgical factors, IOL characteristics, and biological factors. Some of the surgical factors are inaccurate marking for placement, size of the capsulorhexis, or postoperative hypotonia [6, 9]. The biological factors are axial length, large capsular bag, and type of astigmatism [6, 8, 10–14]. Rotation of the IOL may also depend on the IOL characteristics—studies have shown that one-piece C-loop haptic toric IOLs have longer rotation stability [6, 15]. The IOL material is also associated with postoperative rotation—IOLs made with hydrophobic acrylates are more stable than those made with hydrophilic acrylates [6, 16, 17]. Finally, smaller diameter toric IOLs may have higher rotation [6, 8].

The Eyecryl™ toric IOL is a hydrophobic acrylic natural chromophore containing a single-piece lens. The optic size is 6.00 mm (mm), and the overall size is 13.00 mm. It has a refractive index of 1.48 [18]. Previous studies highlighted that the visual and refractive postoperative outcomes of this toric IOL were comparable to those of other IOLs [19]. Based on this literature, we designed the present study. The main objectives of the study are: (1) to evaluate the rotational stability of Eyecryl™ toric IOL and factors associated with rotation 1 month postoperatively and (2) to assess the refractive efficacy in patients implanted with these IOLs and study the factors associated with efficacy in these patients.

Methods

This is a secondary analysis to study the rotation in 121 eyes from 94 patients at a multispecialty eye hospital in Thane district, India.

Study Site and Population

The study has a pre-post design. It was conducted in a private tertiary eye care facility in Thane district, India. Demographic and clinical data from consecutive patients operated on by a single surgeon who were implanted with Eyecryl™ toric IOLs (Biotech Vision Care Pvt Ltd, India) were abstracted and included in the present analysis. The inclusion criteria were: otherwise healthy eyes with senile cataract in the age range of 45–85 years; IOL powers between + 10.00 D and + 30.00 D with IOL implanted in the capsular bag. The exclusion criteria were: eyes with irregular astigmatism, corneal diseases affecting vision, glaucoma, diseases affecting the optic nerve or uveitis, macular diseases or retinopathy, and intra- or postoperative complications (such as posterior capsular rent or cystoid macular edema).

Surgical Procedure Including Rotation Measurement

This procedure is a part of the clinical management of patients who undergo cataract surgery with toric IOLs. The patients were evaluated preoperatively. They underwent a complete anterior and posterior segment with slit-lamp and 90 D lens. The vision was recorded on the Early Treatment Diabetic Retinopathy Study (ETDRS) chart. The IOL power calculation was done using the optical biometry with a Haag-Streit Lenstart LS 900 Optical Biometer (Haag-Streit Group, Switzerland). The toric IOL power calculation was done using the Barrett toric formula on the reference unit of the VERION™ image guided system. The A-constant in the Barrett toric formula was 118.90 with a lens factor of + 1.83 (in accordance with the manufacturer’s recommendations for the implanted IOL model and surgical setup used). The data from Verion were imported to the Digital marker unit of the VERION™ image-guided system.

All patients were operated on by a single surgeon (VS) under topical anesthesia using proparcaine 0.5% twice 15 min before surgery. Betadine 5% eye drops were used twice before the surgery. The draping and eye speculum were placed, and a standard 2.2 mm incision with two sideports was created. The incision was placed at 20° in the right or left eye using the VERION™ image-guided system. A 20° fixed position was chosen as it was easier for a right-handed surgeon to perform phacoemulsification as the incision falls slightly on the right side while operating on the right or the left eye. A standardized capsulorhexis diameter of 5.0 mm was targeted in all cases. This helped to minimize variability as differences in capsulorhexis size and shape are known to influence rotational stability of IOL. The phacoemulsification was done using an Alcon Legion Phacoemulsification machine with parameters according to the cataract grade using a balanced salt solution. Cortex aspiration was done with a coaxial handpiece, and the IOL was injected using a standard injector provided with the lens (Bio Hydro Cartridge BHC 150 and Bio Hydroject Injector) under viscoelastic. The IOL was positioned using the Verion™ Image Guided System. The residual viscoelastic was removed with the irrigation and aspiration system of the phacoemulsification machine. The incisions were sealed by hydrating the wound. The patients were treated with moxifloxacin (0.5%) + dexamethasone (0.1%) eye drops postoperatively. In addition, we included nepafenac (0.1%) and lubricating eye drops (polyethylene glycol [0.4%] and propylene gylcon [0.3%]).

Uncorrected visual acuity (UCVA) and pinhole vision were recorded on day 1 postoperatively. Complete refraction along with subjective acceptance and visual acuity was done at the final follow-up (at day 30 [± 3 days]). Complete slit-lamp examination was done, and the dilated anterior and posterior segment was examined. A slit-lamp imaging system was used to capture the image of the toric IOL with marks. The evaluation was done carefully, and the eyes were at the same level on the chin rest of the slit-lamp imaging system pre- and postoperatively. The image was processed in a Drawboard PDF to measure the exact axis of the toric IOL. This was compared with the axis of placement during surgery. The IOL rotation was calculated by measuring the difference in the clockwise or counterclockwise direction. We abstracted the following data from the clinical records of these patients: (1) demographic data: age, gender; (2) preoperative parameters: right or left eye, refraction (sphere, cylinder, axis, spherical equivalent), corrected distance visual acuity (CDVA), type of astigmatism, axial length, anterior chamber depth (ACD), and aqueous depth (AQD); (3) power of the implanted IOL, axis of the implanted IOL, and predicted residual error; (4) magnitude of postoperative rotation of the IOL (degrees) and direction of rotation; (5) postoperative vision (UDVA and CDVA) and postoperative refraction. The postoperative parameters on day 30 (± 3 days) were used for analysis. The postoperative efficacy was calculated using formula 1: (postoperative cylinder − preoperative anticipated cylinder)/(preoperative cylinder − preoperative anticipated cylinder) × 100 [20].

Statistical Methods

We estimated the means and standard deviation (SDs) for normally distributed linear data. We estimated the median and range or interquartile range (IQR) for data that were not normally distributed. We estimated the proportions for categorical data. The proportions across groups were compared using the chi-square test or Fisher’s exact test for low expected cell counts. The means between two groups were compared using the unpaired t-test. For non-normal data, we used the Wilcoxon-Mann-Whitney rank-sum test for two unpaired groups and Wilcoxon matched-pairs signed rank test for paired data. The vector analysis was done using AstigMATIC software [21]. We also used multivariate logistic regression models to identify the factors associated with rotation and postoperative efficacy. To build a parsimonious model, we did not include highly correlated variables in the same model. There was a significantly high correlation between IOL power and axial length (r = − 0.583), IOL and sphere at baseline (r = 0.441), and IOL and spherical equivalent at baseline (r = 0.431). There was also a significantly high correlation between axial length, anterior chamber depth, and aqueous depth. Thus, in the final model, we included baseline refraction parameters, axial length, type of astigmatism, age, gender, and which eye (right/left). Furthermore, to account for multiple eyes in the same individual, we used mixed effects logistic models. These are better models for correlated data, as is the case in our study; it has been suggested that standard methods may overestimate the effect [22, 23]. The fit of these models was assessed using the Akaike information and Bayesian information criteria [24–26]. P < 0.05 was considered statistically significant.

Data were entered into Ms Excel (© Microsoft, USA) and analysed using Stata Version 17 (© StataCorp, College Station, TX, USA).

The study was approved by the Institutional Ethics Committee of Shree Ramkrishna Netralaya, ref no. SRNEC/ECD/2024/013, 24 July 2024. The study was conducted in accordance with the Declaration of Helsinki 1964 and later amendments. Since this was a retrospective study, a waiver of consent was requested and granted by the Ethics Committee.

Results

The mean (SD) age of these 94 patients was 64.9 (7.2) years; 46% were male (n = 43) and the 54% (n = 51) female. There was no significant difference in the mean age of males and females (64.6 [7.3] years vs 65.3 [7.4] years; p = 0.66). Of the 121 eyes, 61 (50.4%) were right and 60 (49.6%) were left eyes. The median (range) preoperative parameters were as follows: sphere: 0.00 (− 7.00, 5.00) D, cylinder: − 1.02 (− 4.32, − 0.11) D, and spherical equivalent: − 0.50 (− 7.94, 4.39) D. In our study, 19% of eyes had with-the-rule and 81% had against-the-rule astigmatism. The median (IQR) preoperative CDVA was 0.48 (0.18, 0.65) logMAR units. The mean (SD) axial length was 23.1 (0.9) mm, anterior chamber depth was 3.09 (0.34) mm, and aqueous depth was 2.57 (0.33) mm. The mean (SD) of the implanted IOL was 21.6 (2.2) D, and the median (IQR) predicted residual error was 0.09 (0.04, 0.20) D.

There was a significant reduction in sphere, cylinder, and spherical equivalent postoperatively (Table 1). The median (CDVA) had reduced to 0 (0, 0.10) logMAR units; this reduction was statistically significant (p < 0.001). About 86% of the eyes had a postoperative astigmatism of ≤ 0.25 D and another 9% had a postoperative astigmatism of 0.26–0.50 D. In 95% of the eyes, the difference between the attempted and achieved spherical equivalent was within 0.50 D. The refraction and vision changes postoperatively are shown in Fig. 1A–E. The median (IQR) rotation of the IOL on day 30 (± 3 days) was 1 (1, 3) degree and the mean (SD) rotation was 2.31 (2.97) degrees. Most eyes (89%) had a rotation of < 5°; 11% had a rotation of ≥ 5°. There was no significant association between the degrees of rotation ≥ 5° (classified as high rotation) and age, gender, or eye. However, we found that the eyes with high rotation had significantly higher median (IQR) preoperative sphere (more positive side) (1.50 D [0, 1.75] vs 0 D [− 1.0, 0.75]; p = 0.013) and spherical equivalent values (0.96 D [− 0.53, 1.32] vs − 0.50 D [− 1.78, 0.10]; p = 0.02). Although the proportion of eyes with a higher rotation had with-the-rule compared with against-the-rule astigmatism, the difference was not statistically significant (17.4% vs 9.2%; p = 0.17). Similarly, although the mean (SD) of the implanted IOL was higher in eyes with a higher rotation, the difference was not statistically significant (22.7 D [2.3] vs 21.5 D [2.2]; p = 0.07). However, the mean (SD) axial length was significantly lower in eyes with higher rotation (22.6 [0.9] vs 23.2 [0.9]; p = 0.02). Although the mean ACD and AQD were also lower in these eyes as well, the difference was not statistically significant. We present the factors associated with lens rotation in Table 2. We also assessed the rotation in patients in whom bilateral eyes were included. Of these 27 patients, 2 did not have rotation in any eye, 6 had rotation in one eye, and the remaining 19 had rotation in both eyes. In those who had rotation in one eye, five had rotation in the right eye (four clockwise and one counterclockwise) and one had rotation in the left eye (clockwise). In those who had rotation in both eyes, ten had rotation in the same direction (six clockwise and four counterclockwise) in both eyes and the remaining nine had rotation in the opposite direction. The patient with postoperative astigmatism > 0.75 D was informed of repositioning; however, so far, the patient has not come for the procedure.

Table 1.

Pre- and postoperative refractive and vision parameters in 121 eyes implanted with Eyecryl toric intraocular lens, Thane, India

	Preoperative	Postoperative	p value
	Median (range)	Median (range)
Refraction
Sphere (D)	0.00 (− 7.00, 5.00)	0.00 (− 0.50, − 0.75)	0.79
Cylinder (D)	− 1.02 (− 4.32, − 0.11)	0.00 (− 0.75, 1.00)	< 0.001
Axis (D)	88 (1, 176)	0 (0, 170)	< 0.001
Spherical equivalent (D)	− 0.50 (− 7.94, 4.39)	0.00 (− 0.50, 1.13)	< 0.001
Vision
CDVA (logMAR units)	0.48 (0.18, 0.65)	0.00 (0.00, 0.10)	< 0.001

D diopters, CDVA corrected distance visual acuity

Fig. 1.

Graphs reporting on refractive surgery in 121 eyes implanted with Eyecryl toric intraocular lens, Thane, India

Table 2.

Factors associated with postoperative rotation (< 5 and ≥ 5°) in 121 eyes implanted with Eyecryl toric intraocular lens, Thane, India

	All	< 5°	≥ 5°	p value
Total	121 (100.0)	108 (89.3)	13 (10.7)
Age (years)
Mean (SD)	65.2 (7.4)	65.5 (7.4)	62.7 (7.3)	0.21
Gender
Male	54 (44.6)	49 (90.7)	5 (9.3)	0.64
Female	67 (55.4)	59 (88.1)	8 (11.9)
Eye
Right eye	61 (50.4)	55 (90.2)	6 (9.8)	0.75
Left eye	60 (49.6)	53 (88.3)	7 (11.7)
Preoperative refraction Median (IQR)
Sphere (D)	0.00 (− 0.75, 0.75)	0 (− 1.00, 0.75)	1.50 (0.00, 1.75)	0.01
Cylinder (D)	− 1.02 (− 1.33, − 0.72)	− 1.01 (− 1.34, − 0.72)	− 1.09 (− 1.27, − 0.88)	0.77
Axis (D)	88 (73, 99)	88.5 (74, 100)	80 (67, 92)	0.28
Spherical equivalent (D)	− 0.50 (− 1.35, 0.41)	− 0.50 (− 1.78, 0.10)	0.96 (− 0.53, 1.32)	0.02
Type of astigmatism [n (%)]
With-the-rule	23 (19.0)	19 (82.6)	4 (17.4)	0.25
Against-the-rule	98 (81.0)	89 (90.8)	9 (9.2)
Eye parameters [mean (SD)]
Axial length (mm)	23.1 (0.9)	23.2 (0.9)	22.6 (0.9)	0.02
Anterior chamber depth (mm)	3.09 (0.34)	3.10 (0.03)	3.01 (0.22)	0.35
Aqueous depth (mm)	2.57 (0.33)	2.58 (0.34)	2.49 (0.22)	0.39
IOL implanted (D)
Mean (SD)	21.6 (2.2)	21.5 (2.2)	22.7 (2.3)	0.07
Axis IOL implanted (degrees)
Median (IQR)	144 (9, 173)	134.5 (8.5, 171.5)	164 (127, 175)	0.25
Predicted residual error (D)
Median (IQR)	0.09 (0.04, 0.20)	0.08 (0.04, 0.19)	0.17 (0.06, 0.31)	0.27
Type of rotation [n (%)]
No rotation	22 (18.2)	22 (100)	0 (0)	0.14
Clockwise	52 (42.9)	44 (84.6)	8 (15.4)
Counterclockwise	47 (38.8)	42 (89.4)	5 (10.6)
Efficacy
Mean (SD)	85.1 (13.6)	85.8 (13.2)	79.7 (16.1)	0.13
Difference in lines between UDVA and CDVA postoperative
Yes	36 (29.8)	31 (86.1)	5 (13.9)
No	85 (70.3)	77 (90.6)	8 (9.4)	0.47

SD standard deviation, IQR interquartile range, D: diopters, UDVA uncorrected distance visual acuity, CDVA corrected distance visual acuity

The mean (SD) postoperative refractive efficacy was 85.1 (13.6)%. Most patients (83.5%) had > 75% efficacy. Although the efficacy was lower in younger patients, the difference in mean age was not statistically significant (Table 3). There was no association between gender or eye side and refractive efficacy. Similarly, there was no association between preoperative refractive values (sphere, cylinder, axis, and spherical equivalent) and postoperative refractive efficacy. However, efficacy was significantly lower in those who had with-the-rule compared with against-the-rule astigmatism (69.6% vs 86.7%; p = 0.046). We also did not find any association between axial length, ACD, or AQD and refractive efficacy. The median (IQR) predicted residual error was lower in patients who had higher efficacy (0.08 [0.04,0.20] D vs 0.16 [0.08, 0.21] D; p = 0.16); however, this difference was not statistically significant. The proportion of eyes with difference in lines between postoperative UDVA and CDVA was significantly higher in those with low postoperative efficacy (70.0% vs 21.8%; p < 0.001). We present details on factors associated with postoperative efficacy in Table 3. In the vector analysis, we found that in eyes with good efficacy, the targeted induced astigmatism (TIA) was 0.79 D at 85°, the surgically induced astigmatism (SIA) was 0.80 D at 86°, the difference vector was 0.03 D at 33°, and the correction index was 1.00. However, in the eyes with poor refractive efficacy, the TIA was 0.29 D at 93°, the SIA was 0.45 D at 98°, the difference vector was 0.17 D at 16°, and the correction index was 1.35. Thus, in these eyes, there was overcorrection. The graphs are presented in Fig. 2a, b.

Table 3.

Factors associated with refractive efficacy (> 75% and ≤ 75%) in 121 eyes implanted with the Eyecryl toric intraocular lens, Thane, India

	All	=  < 75%	> 75%	p value
Total	121 (100.0)	20 (16.5%)	101 (83.5%)
Age (years)
Mean (SD)	65.2 (7.4)	63.6 (8.8)	65.5 (7.1)	0.32
Gender [n (%)]
Male	54 (44.6)	11 (22.2)	52 (77.8)	0.13
Female	67 (55.4)	8 (11.9)	59 (88.1)
Eye [n (%)]
Right eye	61 (50.4)	10 (16.4)	51 (83.6)	0.97
Left eye	60 (49.6)	10 (16.7)	50 (83.3)
Preoperative refraction Median (IQR)
Sphere (D)	0 (− 0.75, 0.75)	0 (− 1.00, 0.25)	0 (− 0.75, 0.75)	0.40
Cylinder (D)	− 1.02 (− 1.33, − 0.72)	− 0.94 (− 1.14, − 0.43)	− 1.04 (− 1.34, − 0.75)	0.15
Axis (D)	88 (73, 99)	95.5 (69.5, 118.5)	88 (73, 96)	0.18
Spherical equivalent (D)	− 0.50 (− 1.35, 0.41)	− 0.50 (− 1.39, 0.18)	− 0.50 (− 1.35, 0.42)	0.72
Type of astigmatism [n (%)]
With-the-rule	23 (19.0)	7 (30.4)	16 (69.6)	0.046
Against-the-rule	98 (81.0)	13 (13.3)	85 (86.7)
Eye parameters mean (SD)
Axial length (mm)	23.1 (0.9)	23.4 (1.2)	23.1 (0.8)	0.11
Anterior chamber depth (mm)	3.09 (0.34)	3.16 (0.32)	3.08 (0.34)	0.35
Aqueous depth (mm)	2.57 (0.33)	2.64 (0.31)	2.55 (0.34)	0.31
IOL implanted (D)
Mean (SD)	21.6 (2.2)	21.4 (1.7)	21.6 (2.3)	0.68
Axis IOL implanted (degrees)
Median (IQR)	144 (9. 173)	82 (6, 166.5)	147 (10, 173)	0.30
Predicted residual error (D)
Median (IQR)	0.09 (0.04, 0.20)	0.16 (0.08, 0.21)	0.08 (0.04, 0.20)	0.16
Type of rotation [n (%)]
No rotation	22 (18.2)	4 (20.0)	18 (17.8)	0.92
Clockwise	52 (42.9)	9 (45.0)	43 (42.6)
Counterclockwise	47 (38.8)	7 (35.0)	40 (39.6)
Actual rotation (degrees)
Median (IQR)	1 (1, 3)	2 (1, 3)	1 (1, 3)	0.61
Difference in lines between UDVA and CDVA postoperative
Yes	36 (29.8)	14 (70.0)	22 (21.8)	< 0.001
No	85 (70.3)	6 (30.0)	79 (78.2)

SD standard deviation, IQR interquartile range, D diopters, UDVA uncorrected distance visual acuity, CDVA corrected distance visual acuity

Fig. 2.

Vector analysis according to postoperative efficacy (> 75%) (2a) and (≤ 75%) (2b) in 121 eyes implanted with Eyecryl toric intraocular lens, Thane, India

In our study, 75% (n = 91) of the eyes were classified as both low rotation (< 5°) and good efficacy (> 75%); these were categorized as ‘both good’. In addition, 2.5% (n = 3) of eyes were classified as high rotation (≥ 5°) and poor efficacy (≤ 75%). Furthermore, 10 (8.3%) had good efficacy and high rotation, and 17 (14.1%) had poor efficacy and low rotation. All three were categorized as ‘any/both poor’. The parameters in vector analysis for the ‘both good group’ were as follows: TIA: 0.84 D at 85°, SIA: 0.84 D at 86°; difference vector: 0.02 at 31°; correction index: 1.00. These parameters in the ‘any/both poor’ group were as follows: TIA: 0.31 D at 84°, SIA: 0.42 D at 92°; difference vector: 0.15 D at 21°; and correction index: 1.23. The graphs are presented in Fig. 3a, b.

Fig. 3.

Vector analysis according to postoperative rotation and postoperative efficacy in 121 eyes implanted with Eyecryl toric intraocular lens (a both good [efficacy and rotation < 5°] and b any/both poor [efficacy ≤ 75% and rotation ≥ 5°]), Thane, India

In the multivariate mixed effects models, after adjusting for age, gender, eye, axial length, and type of astigmatism, we found that with a 1-unit increase in the preoperative sphere (more positive side), the odds of rotation ≥ 5°were significantly higher (odds ratio [OR] 1.60, 95% confidence intervals [CI] 1.05, 2.42; p = 0.028). However, with a 1-unit increase in the axial length, the odds were significantly lower (OR 0.30, 95% CI 0.09, 0.95; p = 0.04). None of the other variables were associated with rotation of ≥ 5°. In the model with spherical equivalent, we found that with a 1-unit increase in the spherical equivalent (more positive side), the odds of > 5 degree rotation was significantly higher (OR 1.56, 95% CI 1.05, 2.31; p = 0.029). For efficacy, after adjusting for age, gender, baseline refraction, eye, and axial length, we found that eyes with with-the-rule astigmatism had significantly lower odds of efficacy > 75% than against-the-rule astigmatism’ (OR 0.30, 95% CI 0.09, 1.00; p = 0.05).

Discussion

We observed good postoperative stability and refractive efficacy of the Eyecryl toric intraocular lens in these eyes; 75% of the eyes could be classified as having low rotation and good efficacy. The median rotation was low and only 11% of the eyes had a rotation of ≥ 5°. Similarly, the median efficacy was high and 84% had a postoperative efficacy of > 75%. Poor efficacy was associated with preoperative with-the-rule astigmatism, and a rotation ≥ 5° was associated with shorter axial lengths of eyes and higher preoperative spherical values (more positive side). In eyes with poor efficacy and those classified as ‘any/both poor’, the correction index was > 1. Thus, we found that there was overcorrection in these eyes.

Rotational stability is an important concern of operating surgeons post-cataract surgery. A meta-analysis conducted by Li and co-workers found that the mean rotation of the toric IOLs was 2.36° (95% CI 2.08°–2.64°) [27]. This is similar to the mean rotation seen in our data. However, another study by Qiu and co-workers found that the mean rotation was 4.93° and 54% of eyes had a rotation of < 5°; this proportion was considerably lower than seen in our study [28]. In another study, Sugita and co-workers reported that 90% of eyes had a rotation of up to 5° from 1 week to 1 month [29]. Another long-term (1-year follow-up) study by Brar et al. reported a mean rotation of 3.98° at 2 weeks and 4.06° at 1 year; however, they concluded that there was no significant difference between these two values [30]. Multiple factors may be associated with rotation of the IOL:eye-related factors (anatomical characteristics such as axial length), make of the IOL, and operative and postoperative factors (such as accurate marking and placement during surgery) [6]. A study by Li et al. found that rotation of the toric IOL correlated positively with axial length [9]. Another study by Osawa et al., however, did not find any significant correlation between rotation and axial length [31]. Shi and co-workers found that there was no significant difference in mean rotation of monofocal and multifocal IOL groups in eyes with axial length < 26 mm or those with an axial length of ≥ 26 mm [32]. It has been suggested that a rotation of up to 5° may be clinically relevant [27, 33]. In our study, we found that the eyes with ≥ 5° rotation had significantly shorter axial lengths. However, there was no significant association between rotation and anterior chamber depth, a finding also seen in the study by Li et al. [9]. Although we studied the nature of rotation including both eyes, this is a potential area for future studies. Key findings from the included studies are presented in Table 4.

Table 4.

Findings from select studies included in the discussion section

Authors, reference no.	Journal, region	Key findings
Li ES et al. [27]	Journal of Cataract and Refractive Surgery, Systematic Review and meta-analysis	51 studies included for meta-analysis Included various types of IOLs Mean pooled absolute rotation was 2.36° (95% confidence intervals, 2.08 to 2.64) Sub group analysis was also done based on lens model and haptic design
Qiu X et al. [28]	Journal of Ophthalmology China	57 eyes from 57 patients Mean (SD) toric IOL rotation was 4.93 (3.02) degrees 54.39% of eyes had rotation < 5° 5.26% of eyes (3 eyes) had rotation > 10° 47.3% had residual astigmatism < 0.75 D
Sugita I et al. [29]	BMC Ophthalmology, Japan	64 eyes of 53 patients Mean (SD) absolute rotation was 2.2 (1.7) degrees from 1 week to 1 month; 90% of eyes had a rotation < 5°, 10% had a rotation of 5°–10° Mean (SD) absolute rotation was 2.1 (2.0) degrees from 3 to 6 months; 87% of eyes had a rotation < 5°, 13% had a rotation of 5°–10° Mean (SD) absolute rotation was 3.1 (2.9) degrees from 6 to 12 months; 77% of eyes had a rotation < 5°, 18% of eyes had a rotation of 5°–10°, and 5% had rotation > 10°
Brar S et al. [30]	Indian Journal of Ophthalmology, India	50 eyes from 39 patients Mean (SD) absolute rotation was 3.98 (2.15) degrees at 2 weeks Mean (SD) absolute rotation at 12 months was 4.06 (2.15) degrees; the difference between 2 weeks and 12 months was not statistically significant At 12 months, 84% of eyes had rotation < 5° and 16% between 6°–10°
Li S et al. [9]	Eye, China	102 eyes from 87 patients Mean (SD) at 3 months was 4.83 (3.65) degrees 68.63% of eyes had a rotation < 5°, 26.47% had rotation between 5°–10°, and remaining 5% had rotation > 10° Significant association between IOL rotation and axial length (r = 0.259, p = 0.009) No significant association between IOL rotation and anterior chamber depth (r = 0.011, p = 0.94)
Osawa R et al. [31]	PLOS One, Japan	193 eyes from 165 patients Presented comparison of four different lenses (AcrySof, TECHNIS, HOYA 355, and HOYA XY-1) Axis misalignment < 10° was seen in 82.0% of eyes with AcrySof, 64.7% in TECHNIS, 54.3% in HOYA 355, and 89.1% in HOYA XY-1 No significant correlation between axial length and misalignment at 1 month
Shi R et al. [32]	BMC Ophthalmology, China	422 eyes from 422 patients and propensity score matching of 126 in the monofocal group with 126 from the multifocal group Higher mean (SD) rotation in the monofocal group compared with the multifocal group 5.40 (7.77) in the monofocal group vs 3.52 (3.54) in the multifocal group ( p = 0.019) There was no significant difference in the mean rotation in eyes with axial length < 26 mm or ≥ 26 mm

In our study, we found that the mean refractive efficacy was high. Previous studies have also found a high refractive efficacy with toric IOLs. For instance, Wang and co-workers found that the mean absolute error was 0.39 D and about 72% of the eyes had an absolute error of 0.50 D postoperatively [34]. Ackerman et al. found the refractive efficacy to be high for the Panoptix toric intraocular lens. They reported that 61% of eyes had UDVA ≥ 20/20, and 94% had a refractive astigmatism of ≤ 0.5 D [35]. Hwang et al. compared rotation and efficacy in two types of toric IOLs; they found that the mean postoperative rotation was significantly lower in the Technis Eyehance Toric II compared with Technis TIOL and satisfaction was better in the former group compared with latter [36]. In our study, the efficacy was high and was associated with the type of astigmatism—against-the-rule had higher efficacy than with-the-rule astigmatism. A study by Hasegawa et al. however, found that UDVA was worse ins against-the-rule and oblique astigmatism in pseudophakic eyes [37]. In our study, the correction index was 1.00 and the difference vector was very low (0.02) in eyes that had lower rotation and higher refractive efficacy. A similar finding was reported by Novacek and co-workers. They found that 94% of eyes had a residual cylinder up to 0.50 D and mean (SD) rotation was 1.42 (1.89) degrees. They reported the correction index to be 0.96 with a difference vector of 0.17; thus, the correction index and difference vector were associated with good outcomes in these surgeries [38].

There are some potential limitations of our study. We only assessed the rotation at 1 month post-surgery. Some earlier authors have studied rotational changes for up to 6 months or even a year [30, 38]. However, other authors have reported that the most important rotation period was within 24 h post-surgery, and there was little rotation from the first to the third month of the postoperative period [6, 32, 39]. This is a limitation of the study. For rotational assessment, we measured the rotation with a slit-lamp and compared it to the axis of placement during surgery rather than a fixed anatomical landmark; this may be a potential limitation of the study. The mean (SD) axial length at baseline was 23.1 (0.9) mm; this is a very narrow range, and we did not have many shorter eyes. This may limit the generalizability of our results. Nonetheless, in our group we found that an increase in the axial length was significantly associated with a lower odds of rotation of ≥ 5°. Furthermore, most of the patients in our study had a preoperative astigmatism of ≤ 2 D. This also may limit the generalizability of our study. Some other factors such as the material and design of the lens, surgical factors, and postoperative physical activity may be associated with rotational stability [27]. We only used the type of lens, and only one surgeon performed all the surgeries. The same person also did the pre- and postoperative assessment of the lens position. Thus, we minimized the variability in procedural and post-procedural factors. However, we did not have the exact level of physical activity of individuals; this could be another potential limitation of the study.

Conclusion

We found that the Eyecryl™ toric intraocular lens had low rotation even after 1 month post-surgery. The mean rotation was 2.31°, and the median rotation was 1°. Most (84%) of the eyes had a rotation of < 5°; higher rotation was associated with a higher preoperative sphere (more positive side) and shorter axial length (short eyes). Postoperative efficacy was higher in most eyes; this was associated with against-the-rule astigmatism. Eyes that had either lower efficacy/high rotation or both showed overcorrection on vector analysis. However, in general, Eyecryl™ toric IOL was stable even 1 month after cataract surgery. Thus, refraction stability and optimal visual acuity were probably attained post-surgery. This will ensure consistent refractive correction.

Author Contributions

Vijay Shetty and Nitin Deshpande conceptualized the study. They also provided comments on the manuscript, including interpretation of findings, and helped edit the manuscript. Prajakta Deshpande helped in overall supervision of the study, interpretation of findings, and drafting of the manuscript. Jekin Choubisa, Akshay Chavan, and Amruta Pradhan helped with data cleaning, data analysis, and manuscript preparation. They also helped the other members with data collection and with the preparation of tables and figures with other team members. Seeba John and Alfin Shaji helped with data collection and data cleaning. They also helped with the preparation of the tables and figures for the manuscript. Maninder Singh Setia analysed the data and helped draft the manuscript with other team members.

Funding

The authors received funds for publication of this article from Biotech Healthcare, India. The funder has provided funds for Rapid Service Fee.

Data Availability

Data can be made available after reasonable request to the corresponding author or the Ethics Committee at ethics@shreeramkrishnanetralaya.com. The data will be made available after administrative and ethical approval and in accordance with local data protection.

Declarations

Conflict of Interest

The authors declare that they have no conflict of interest. Vijay Shetty, Nitin Deshpande, Prajakta Deshpande, Jekin Choubisa, Akshay Chavan, Alfin Shaji, Seeba John, Amruta Pradhan, and Maninder Singh Setia do not have any conflicts of interest.

Ethical Approval

The study was approved by the Institutional Ethics Committee of Shree Ramkrishna Netralaya ref no. SRNEC/ECD/2024/013, 24 July 2024. The study was conducted in accordance with the Declaration of Helsinki 1964 and later amendments. Since this was a retrospective study, a waiver on consent was requested and granted by the Ethics Committee.