Effect of image-guided systems in phacoemulsification with intraocular lens (IOL) implantation: a systematic review and meta-analysis

Abstract

Purpose

To explore the effect of image-guided systems in phacoemulsification with intraocular lens (IOL) implantation.

Methods

We searched Pubmed, Embase and China National Knowledge Infrastructure (inception to January 20, 2021). Two researchers extracted data and assessed paper quality independently. Uncorrected distance visual acuity (UDVA) before and after surgery, best corrected visual acuity (BCVA) before and after surgery, preoperative cylinder, postoperative residual refractive cylinder, postoperative corneal cylinder, IOL misalignment, and intraocular pressure (IOP) were compared.

Results

We included 14 studies with 885 cataract eyes. All data were performed using Review Manager 5.3 (RevMan 5.3) (https://revman.cochrane.org/). Cases of all preoperative outcomes showed no significant difference between image-guided group and manual group. There was no significant difference in postoperative UDVA (Standard mean difference (SMD: −0.11, 95% CI: −0.32 to 0.11, I² = 59%, p = 0.33)), BCVA (SMD: 0.03, 95% CI: −0.12 to 0.18, I² = 36%, p = 0.72), corneal cylinder (Weighted mean difference WMD: 0.13, 95% CI: −0.06 to −0.32, I² = 0%, p = 0.17), IOP (WMD: −0.37, 95% CI: −1.36 to −0.62, I² = 9%, p = 0.46) between two groups. There was less residual refractive cylinder in image-guided group than in manual group (WMD: −0.20, 95% CI: −0.26 to −0.14, I² = 59%, p＜0.00001). It is more accurate in IOL alignment when combined with image-guided systems (WMD: −1.20, 95% CI: −1.43 to −0.96, I² = 14%, p < 0.00001).

Conclusion

Image-guided systems can improve the effect in phacoemulsification with intraocular lens (IOL) implantation.

Keywords:

• Cataract
• image-guided
• IOL implantation
• systematic review

Introduction

Cataract is a group of opacity in the lens as a common feature of the disease, one of the major blinding eye diseases in the world, and surgery is the main treatment. 34.7%–47.3% of cataract patients had corneal cylinder≥ 1.0 D, which seriously affected the visual function after refractive surgery [1]. Intraocular lens (IOL) implantation is a safe and effective modality in such cases to help treat cataract at the same time of correcting regular cylinder of cornea [2,3]. Accurate measurement of preoperative corneal cylinder, IOL axis alignment along the desired corneal meridian and intraoperative positioning are necessary. One degree of misalignment results in 3.3% decrease in the correction of cylinder, and 30° misalignment effectively increase the amount of preoperative cylinder [4–6].

Manual marking has been widely used, such as bubble, direct vision in the slitlamp, ink, pendulum and tonometer [7]. With the development of digitization, imaged-guided systems are gradually used in cataract surgery, such as the Callisto Eye with ZAlign (Carl Zeiss Meditec AG, Jena, Germany), the TrueGuide software (TrueVision 3 D Surgical, Inc., Santa Barbara, CA), and the VERION Digital Marker (Alcon Laboratories, Ft. Worth, TX) [8–10]. Compared with manual marking, the image-guided systems have the functions of intraoperative projection, incision positioning, assisted capsulorhexis trajectory and toric IOL positioning.

However, which method is the better remains controversial. Three studies [11–13] suggested that image-guided systems have a significant effect on improving the vision acuity in phacoemulsification with IOL implantation. Five studies [12–16] suggested that there were less residual refractive cylinder in image-guided marking group than in manual marking group while other four studies [17–20] didn’t. In addition to IOL misalignment, strong evidence indicates that it was more accurate when combined with image-guided systems [13,15–22]. And two studies [14,23] suggested accurate manual marking and digital marking were equally effective guides for toric IOL alignment intraoperatively.

The aim of this systematic review and meta-analysis is to identify, evaluate and summarize the available evidence to determine if image-guided systems have an overall significant improvement in phacoemulsification with IOL implantation.

Material and methods

Literature search

The meta-analysis was designed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. The Pubmed, Embase and China National Knowledge Infrastructure (inception to January 20, 2021) databases were searched for cohort studies. The search terms were “(cataract OR IOL) AND (digital OR guided OR verion)” and they were applied without language restrictions. Studies from other sources were added if they met the inclusion criteria.

Study eligibility and selection

The following studies were included:

• Participants’ eyes with corneal cylinder were at least 1.00 D,

• Participants’ eyes had cataract surgery and toric IOL implantation,

• Surgery used both image-guided and manual marking techniques for toric alignment.

Studies were excluded if

• Patients were with ocular pathology such as preexisting glaucoma, retinal pathology, small pupils, corneal opacity, ocular surface disorders, or systemsic conditions, or medication use expected to result in a postoperative visual acuity worse than 20/30,

• Patients currently used contact lenses or had previous refractive surgery or conditions with the potential to result in a complicated cataract surgery,

• Eyes were with intraoperative complications such as extension of capsulorhexis, posterior capsular rent, vitreous loss, and nucleus drop.

Each study was initially screened by two investigators (Yang Su, Guo-Guo Yi) based on their titles and abstracts, and then by reading the full text according to the inclusion and exclusion criteria, with any disagreement resolved by consultation with a third investigator (Min Fu).

Quality assessment

The quality of the papers was evaluated using the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool (http://www.bristol.ac.uk/population-health-sciences/projects/quadas/) from the following 4 aspects: patient selection, index text, reference standard, flow and timing.

Data extraction

We extracted the following data from each study:

• Study characteristics, including the first author, publication year, study design, surgery method, number of patients, age,

• Outcome measures, including uncorrected distance visual acuity (UDVA) before and after surgery, best corrected visual acuity(BCVA) before and after surgery, preoperative cylinder, postoperative residual refractive cylinder, postoperative corneal cylinder, IOL misalignment, and intraocular pressure(IOP).

Data synthesis and statistical analysis

All measures we extracted (UDVA, BCVA, cylinder, IOL misalignment, IOP) were reported as the mean ± SD. UDVA and BCVA before and after surgery were reported by decimal counting in the two studies [12,13] and log MAR in other studies, therefore analyzed by the standard mean difference (SMD). Cylinder, IOL misalignment and IOP were analyzed by weighted mean difference (WMD). Subgroup analyses according to the follow-up time were performed by grouping studies that used the same tool. p < 0.05 was considered statistically significant.

The I² statistic was assessed to measure the degree of inconsistency and quantify the effect of heterogeneity in the results of the trials. Outcomes from the selected trials were combined with a fixed-effect model if I² was less than 50%. The random-effect model was applied if I² was greater than 50%, which indicates substantial heterogeneity. Sensitivity analyses were performed first by removing the studies with the highest risk of bias and then by removing one study at a time to verify whether the results would change. All statistical analyses were conducted using Review Manager Software (Version 5.3, The Nordic Cochrane Center, The Cochrane Collaboration, 2014, Copenhagen) (https://revman.cochrane.org/).

Results

Characteristics of studies

A total of 3245 articles were initially identified, with 1129 of these duplicates, leaving 2116. Following the screening of titles and abstracts, 40 articles were retrieved for full-text review. 14 articles [11–24] that met the inclusion criteria were finally included. The flowchart (Figure 1) showed the literature search process.

Figure 1. Flow chart of paper selection.

The characteristics of the included studies are showed in Table 1. The studies were published between 2014 and 2020 and included a total number of 885 eyes (469 in image-guided group and 416 in manual group) in phacoemulsification with IOL implantation. The duration of the follow up ranged from 1 day to 3 months.

Table 1. Characteristics of included studies.

Study ID	Eyes/Patients	Study design	Surgery	Group
				Image-guided	No.	Age	Manual	No.	Age
Shreesha K Kodavoor 2020	61/50	Pro	Phaco + IOL	VERION	30	62.68 ± 5.50	Bubble	31	63.04 ± 9.77
Wan Jiayu 2020	40/-	Pro	Phaco + IOL	Callisto Eye	20	63.85 ± 12.09	Slitlamp	20	68.15 ± 7.85
Xuan Zhongning 2019	76/76	Pro	Phaco + IOL	VERION	38	71.00 ± 5.60	Slitlamp	38	70.90 ± 4.50
Cao Yingxue 2019	108/108	Pro	Phaco + IOL	Callisto Eye	76	67.42 ± 10.21	–	32	59.40 ± 7.90
Jeewan S Titiyal 2018	80/80	Pro	Phaco + IOL	Callisto Eye	40	60.10 ± 10.30	Bubble	40	63.10 ± 13.70
Zhang Bei 2018	80/80	Pro	Phaco + IOL	VERION	40	56.40 ± 8.00	Slitlamp	40	59.40 ± 7.90
Wang Xiaoming 2018	75/75	Petro	Phaco + IOL	VERION	42	62.98 ± 9.36	Slitlamp	33	61.52 ± 9.88
Wolfgang J. Mayer 2017	57/29	Pro	Phaco + IOL	Callisto Eye	29	–	Bubble	28	–
Valentijn S.C. Webers 2017	35/24	Pro	Phaco + IOL	VERION	18	–	Ink	17	–
Wang Qing 2017	60/60	Pro	Phaco + IOL	VERION	30	67.78 ± 11.93	Slitlamp	30	69.02 ± 13.63
Zhang Ziyin 2017	61/61	Pro	Phaco + IOL	Callisto Eye	30	68.72 ± 7.12	–	31	68.93 ± 7.59
Wu Lei 2017	40/40	Petro	Phaco + IOL	VERION	20	72.60 ± 4.50	Slitlamp	20	68.60 ± 13.20
Ildamaris Montes de Oca 2016	26/26	Petro	Phaco + IOL	the Truevision 3-dimensional	26	–	Ink	26	–
Abdel Hamid Elhof 2015	60/60	Pro	Phaco + IOL	VERION	30	49.50 ± 11.40	Pendulum	30	52.00 ± 13.00

Table 2 and Figure 2 showed the risk of bias and applicability concerns in all. Four studies [15,16,22,23] did not include information on whether they enrolled a consecutive series of patients and were case control studies, resulting in high degrees of risk for bias in patient selection. How and who measured refraction has the potential to bias the study endpoint. Few studies mention whether the index test results were interpreted without knowledge of the results of the reference standard and whether the reference standard results were interpreted without knowledge of the results of the index test.

Figure 2. Methodological quality graph.

Table 2. Quality assessment.

Study	Patient selection					Index test(s)				Reference standard				Flow and timing
	1	2	3	Risk	Concern	4	5	Risk	Concern	6	7	Risk	Concern	8	9–10	11	Risk
Shreesha K Kodavoor 2020	Yes	Yes	No	HR	LC	Unclear	Yes	UR	UC	Yes	Unclear	UR	UC	Yes	Yes	Yes	LR
Wan Jiayu 2020	Yes	Yes	Yes	LR	LC	Unclear	Yes	UR	UC	Yes	Unclear	UR	UC	Yes	Yes	Yes	LR
Xuan Zhongning 2019	Yes	Yes	No	HR	LC	Unclear	Yes	UR	UC	Yes	Unclear	UR	UC	Unclear	Yes	Yes	UR
Cao Yingxue 2019	Yes	Yes	Unclear	UR	UC	Unclear	Yes	UR	UC	Yes	Unclear	UR	UC	Yes	Yes	Yes	LR
Jeewan S Titiyal 2018	No	Yes	Unclear	UR	UC	Unclear	Yes	UR	UC	Yes	Unclear	UR	UC	Unclear	Yes	Yes	UR
Zhang Bei 2018	Yes	Yes	Yes	LR	LC	Unclear	Yes	UR	UC	Yes	Unclear	UR	UC	Unclear	Yes	Yes	UR
Wang Xiaoming 2018	No	Yes	Unclear	HR	HC	Yes	Yes	LR	LC	Yes	Yes	LR	LC	Yes	Yes	Yes	LR
Wolfgang J. Mayer 2017	Yes	Yes	Unclear	UR	UC	Unclear	Yes	UR	UC	Yes	Unclear	UR	UC	Unclear	Yes	Yes	UR
Valentijn S.C. Webers 2017	Yes	Yes	Yes	LR	LC	Unclear	Yes	UR	UC	Yes	Unclear	UR	UC	Unclear	Yes	Unclear	UR
Wang Qing 2017	Yes	Yes	Unclear	UR	UC	Unclear	Yes	UR	UC	Yes	Unclear	UR	UC	Yes	Yes	Yes	LR
Zhang Ziyin 2017	Yes	Yes	Unclear	UR	UC	Unclear	Yes	UR	UC	Yes	Unclear	UR	UC	Unclear	Yes	Yes	UR
Wu Lei 2017	No	Yes	Unclear	HR	HC	Yes	Yes	LR	LC	Yes	Yes	LR	LC	Yes	Yes	Yes	LR
Ildamaris Montes de Oca 2016	No	Yes	Unclear	HR	HC	Yes	Yes	LR	LC	Yes	Yes	LR	LC	Unclear	Yes	Yes	UR
Abdel Hamid Elhof 2015	Yes	Yes	Yes	LR	LC	Unclear	Yes	UR	UC	Yes	Unclear	UR	UC	No	Yes	Yes	HR

LR: low risk; HR: high risk; UR: unclear risk; LC: low concern; HC: high concern; UC: unclear concern.

The Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool evaluates 11 criteria 1. Was a consecutive or random sample of patients enrolled? 2. Was a case-control design avoided? 3. Did the study avoid inappropriate exclusions? 4. Were the index test results interpreted without knowledge of the results of the reference standard? 5. If a threshold was used, was it pre-specified? 6. Is the reference standard likely to correctly classify the target condition? 7. Were the reference standard results interpreted without knowledge of the results of the index test? 8. Was there an appropriate interval between index test(s) and reference standard? 9. Did all patients receive a reference standard? 10. Did patients receive the same reference standard? 11. Were all patients included in the analysis?

Preoperative outcomes

There were six studies [12–16,21] reported preoperative UDVA, eight studies [11–15,18,21,22] reported preoperative BCVA, and twelve studies [11–15,17–22,24] reported preoperative cylinder. There was no significant difference in UDVA (SMD: 0.25, 95% CI: −0.21 to 0.70, I² = 77%, p = 0.29), BCVA (SMD: −0.10, 95% CI: −0.27 to 0.08, I² = 37%, p = 0.28) and cylinder (WMD: 0.03, 95% CI: −0.05 to 0.10, I² = 19%, p = 0.48) between image-guided group and manual group (Figure 3A–C).

Figure 3. (A) Preoperative UDVA; (B) Preoperative BCVA; (C) Preoperative cylinder.

Postoperative outcomes

UDVA

It is divided into five subgroups according to the follow-up time in UDVA. There were two studies [12,15] reported in the first subgroup (1 day after surgery) (SMD: −0.01, 95% CI: −0.66 to 0.64, I² = 66%, p = 0.97), two studies [14,15] in the second subgroup (1 week after surgery) (SMD: 0.02, 95% CI: −0.67 to 0.72, I² = 67%, p = 0.95), two studies [12,15] in the third subgroup (1 month after surgery) (SMD: 0.04, 95% CI: −0.41 to 0.48, I² = 31%, p = 0.87), seven studies [11–14,16,18,20] in the fourth subgroup (3 months after surgery) (SMD: −0.14, 95% CI: −0.54 to 0.27, I² = 75%, p = 0.50), and two studies [17,21] in the fifth subgroup (the follow-up time unknown) (SMD: −0.36, 95% CI: −0.72 to 0.01, I² = 0%, p = 0.06). There was no significant difference in each subgroup respectively and in the total group (SMD: −0.11, 95% CI: −0.32 to 0.11, I² = 59%, p = 0.33) between image-guided group and manual group (Figure 4)

Figure 4. Postoperative UDVA in five subgroups respectively and in a total group.

BCVA

It is divided into four subgroups according to the follow-up time in BCVA. There were two studies [12,15] reported in the first subgroup (1 day after surgery) (SMD: −0.23, 95% CI: −0.59 to 0.13, I² = 0%, p = 0.21), two studies [12,14] in the second subgroup (1 week after surgery) (SMD: 0.15, 95% CI: −0.24 to 0.54, I² = 0%, p = 0.46), two studies [12,15] in the third subgroup (1 month after surgery) (SMD: 0.24, 95% CI: −0.52 to 1.00, I² = 74%, p = 0.54), six studies [11–14,18,21] in the fourth subgroup (3 months after surgery) (SMD: 0.05, 95% CI: −0.17 to 0.26, I² = 49%, p = 0.68). There was no significant difference in each subgroup respectively and in the total group (SMD: 0.03, 95% CI: −0.12 to 0.18, I² = 36%, p = 0.72) between image-guided group and manual group (Figure 5).

Figure 5. Postoperative BCVA in four subgroups respectively and in a total group.

Residual refractive cylinder

It was divided into four subgroups according to the follow-up time. There were three studies [12,14,24] reported in the first subgroup (1 week after surgery) (WMD: −0.23, 95% CI: −0.36 to −0.11, I² = 0%, p = 0.0003), two studies [12,24] in the second subgroup (1 month after surgery) (WMD: −0.26, 95% CI: −0.37 to −0.14, I² = 0%, p < 0.00001), eight studies [12–14,16,18–20,24] in the third subgroup (3 months after surgery) (WMD: −0.19, 95% CI: −0.29 to −0.10, I² = 73%, p = 0.0001), and two studies [15,17] in the fourth subgroup (the follow-up time unknown) (WMD: −0.15, 95% CI: −0.34 to 0.03, I² = 65%, p = 0.10). Although the difference was not statistically significant in the fourth group, it was statistically significant in other three subgroups respectively and in the total group (WMD: −0.20, 95% CI: −0.26 to −0.14, I² = 59%, p < 0.00001) between image-guided group and manual group (Figure 6).

Figure 6. Residual refractive cylinder in four subgroups respectively and in a total group.

Corneal cylinder

Two studies [15,19] showed a comparison of corneal cylinder between the image-guided group and manual group. The difference was not statistically significant (WMD: 0.13, 95% CI: −0.06 to −0.32, I² = 0%, p = 0.17) (Figure 7).

Figure 7. Postoperative corneal cylinder.

IOL misalignment

It is divided into five subgroups according to the follow-up time. There were four studies [15,18,19,22] reported in the first subgroup (1 day after surgery) (WMD: −1.23, 95% CI: −1.78 to −0.68, I² = 13%, p < 0.0001), five studies [12,14,18,22,23] in the second subgroup (1 week after surgery) (WMD: −0.74, 95% CI: −1.26 to −0.22, I² = 13%, p = 0.005), four studies [12,15,18,22] in the third subgroup (1 month after surgery) (WMD: −1.21, 95% CI: −1.73 to −0.70, I² = 2%, p < 0.00001), seven studies [12,14,16,18–20,22] in the fourth subgroup (3 months after surgery) (WMD: −1.35, 95% CI: −1.76 to −0.93, I² = 29%, p< 0.00001), two studies [17,21] in the fifth subgroup (the follow-up time unknown) (WMD: −1.64, 95% CI: −2.46 to −0.83, I² = 0%, p< 0.0001). There was no significant difference in each subgroup respectively and in the total group (WMD: −1.20, 95% CI: −1.43 to −0.96, I² = 14%, p<0.00001) between image-guided group and manual group (Figure 8).

Figure 8. IOL misalignment in five subgroups respectively and in a total group.

IOP

Two studies [12,15] showed a comparison of IOP between the image-guided group and manual group. The difference was not statistically significant (WMD: −0.37, 95% CI: −1.36 to −0.62, I² = 9%, p = 0.46) (Figure 9).

Figure 9. Postoperative IOP.

Heterogeneity analysis

Postoperative UDVA (I² = 59%) and residual refractive cylinder (I² = 59%) have high heterogeneity. After excluding Cao and Wu’s studies [11,16] in UDVA, the heterogeneity significantly decreased in the fourth group (3 months after surgery) (SMD: 0.13, 95% CI: −0.11 to 0.37, I² = 26%, p = 0.27) and in the total group (SMD: −0.02, 95% CI: −0.17 to 0.12, I² = 34%, p = 0.76). The difference was still not statistically significant. After excluding Wu’s study [16], the heterogeneity significantly decreased in the fourth group (3 months after surgery) (WMD: −0.13, 95% CI: −0.17 to −0.08, I² = 33%, p<0.00001) and in the total group (WMD: −0.15, 95% CI: −0.19 to −0.11, I² = 33%, p<0.00001), the result did not change and it was still statistically significant. Other outcomes all have low heterogeneity.

Discussion

With the development of digital technology, cataract patients have high expectations for the curative effect of surgery. Two main factors are essential: the insertion of a correct power IOL due to accurate keratometry [25,26] and accurate alignment of toric IOL. Image-guided systems can not only provide keratometry (K) measurements, but provide digital image guidance for toric IOL alignment without preoperative manual marking. Image-guided systems were designed to help the surgeon with a preoperative planning of the size and location of the surgical incisions and capsulorrhexis, as well as correct placement of IOLs and accurate alignment of toric IOLs. These systems give the opportunity to the surgeon to see the aforementioned measurements intraoperatively in real time [26,27].

This meta-analysis is to explore the effect of image-guided systems in phacoemulsification with IOL implantation. The difference was not statistically significant in all preoperative measurements we extracted (UDVA, CDVA and cylinder), making our postoperative results more accurate credible. 3 methods (Verion, Alcon Laboratories, Inc.; Callisto Eye Systems, Carl Zeiss Meditec AG; and TrueVision 3 D, TrueVision Systems, Inc.) were included in the image-guided group and 4 methods (bubble, pendulum, direct visualization by the silmlamp and ink) in the manual group, increasing the heterogeneity. And we divided postoperative outcomes (UDVA, CDVA, residual refractive cylinder and IOL misalignment) into different subgroups according to the following-up time ranging from 1 day to 3 months, decreasing the heterogeneity [28,29]. In the analyses above, residual refractive cylinder had high heterogeneity. After excluding the study [16] of high heterogeneity, the result did not change and it was still statistically significant, which mean our results were credible.

The results of this meta-analysis suggest that image-guided systems can improve the effect in cataract surgery. Combined with image-guided systems, less residual refractive cylinder and more accurate IOL alignment were in phacoemulsification with IOL implantation. Although no significantly difference in postoperative UDVA and BCVA, the vision acuity in image-guided groups were commonly better than in manual groups. Larger-sample experiments should be performed to verify if image-guided systems can really have a significant effect on improving vision acuity. It was not statistically significant in postoperative UDVA, BCVA, corneal cylinder and IOP, suggesting that manual marking can also achieve high surgical efficacy in the condition of limited economic.

Also, there were some limitations in this meta-analysis: (1) the sample size we included was small, and a large number of studies were still needed; (2) the digital systems in image-guided groups (Verion, Alcon Laboratories, Inc.; Callisto Eye Systems, Carl Zeiss Meditec AG; and TrueVision 3 D, TrueVision Systems, Inc.) and the manual marking methods (bubble, pendulum, direct visualization by the silmlamp and ink) in manual groups were different; (3) most of the included studies did not adopt blind method and allocation concealment, and there was the possibility of selective bias, implementation bias and measurement bias; (4) the number of some indicators included (postoperative corneal cylinder and IOP) was small, and the statistical efficiency may be insufficient. In conclusion, image-guided systems can improve the effect in phacoemulsification with intraocular lens (IOL) implantation. And more high-quality and large-scale clinical studies are needed.

Ethics statement

This article is based on previously published studies and does not contain any studies with human participants performed by any of the authors.

Author contributions

All named authors meet the International Committee of Medical Journal Editors (ICMJE) criteria for authorship for this article, take responsibility for the integrity of the work as a whole, and have given their approval for this version to be published.

Disclosure statement

Yang Su, Min Fu, Yifan Chen, Rui Qiao and Guo-Guo Yi declare no personal, financial, commercial, or academic conflicts of interest, and none of the authors have a financial or proprietary interest in a product, method, or material in this paper.

Data availability statement

The datasets generated during and/or analyses during the current study are available from the corresponding author on reasonable request.

Additional information

Funding

This study was supported by Natural Science Foundation of Beijing Municipality (Grant No. 7192222) and National Natural Science Foundation of China (Grant No. 82072352).