Article Text
PDF
Abstract
Purpose To analyse the occurrence and potential causes of lens capsule-related complications during femtosecond laser-assisted cataract surgery (FLACS).
Methods This prospective consecutive cohort study included the first 1600 eyes (from 1140 consecutive patients) who received FLACS performed by the same surgeon from May 2015 to December 2018. The potential causes and characteristic signs of capsulotomy-related complications, including incomplete capsulotomies and radial anterior capsule (AC) tears, were summarised based on the agreement of two ophthalmologists after they analysed the surgical videos. Subgroup analysis was conducted to characterise the capsulotomy learning curve.
Results Of the 1600 eyes, 52 (3.25%) had incomplete capsulotomies and 22 (1.38%) had radial AC tears. The most common causes of incomplete capsulotomies were eye tilt (16 eyes, 30.77%), air bubbles or ocular secretions at the interface (14 eyes, 26.92%) and white cataracts (7 eyes, 13.46%). Additionally, 54.55% (12/22) of AC tears were due to incomplete capsulotomy and secondary capsulorhexis. A significant difference was noted between the first 200 eyes and subsequent groups in terms of the incidence of incomplete capsulotomies. No difference was observed in the incidence of AC tears after the initial 100 procedures.
Conclusion The most common causes of incomplete capsulotomies were eye tilt and air bubbles or ocular secretions at the interface. Secondary capsulorhexis after incomplete capsulotomy is the main risk factor for AC tears. There was a steep learning curve for laser capsulotomy in the first 100 operated eyes, as evidenced by the higher complication rate, but this stabilised after 200 procedures.
- lens and zonules
- treatment lasers
- treatment surgery
Data availability statement
Data are available upon reasonable request. Not applicable.
Statistics from Altmetric.com
Introduction
Femtosecond laser-assisted cataract surgery (FLACS) has gained popularity in recent years due to its many advantages.1–3 One of the greatest benefits of FLACS is its ability to achieve a consistently well-sized and properly shaped capsulotomy with greater precision,4 which improves surgical safety and results in better intraocular lens (IOL) centration. This is particularly beneficial when implanting premium IOLs, such as toric5 and multifocal IOLs,6 which require precise positioning for optimal clinical performance.
However, when compared with manual capsulorhexis, FLACS also has some unique complications, mainly comprising incomplete capsulotomies and radial anterior capsule (AC) tears.7–9 These complications, if not recognised early and properly treated, may lead to serious conditions such as IOL decentration or tilt, posterior capsule rupture or even nucleus dislocation and vitreous prolapse. Therefore, during FLACS, capsulotomy-related complications, especially AC tears, are the most concerning and tricky issue for surgeons.
Current studies on laser capsulotomy are mostly limited to incidence reports; hence, there is a lack of specific analysis of the causes of complications. From this point of view, this study aims to analyse the potential causes and characteristic signs of capsulotomy-related complications during FLACS by studying videos of the surgeries performed. We also evaluate the surgeon’s learning curve in laser capsulotomy and summarise our experiences in preventing complications.
Materials and methods
Patients
This prospective consecutive cohort study included all patients who underwent LenSx FLACS (LenSx, Alcon Laboratories, Elkridge, Maryland, USA) performed by the same surgeon (KY) at the Second Affiliated Hospital of the School of Medicine, Zhejiang University, Hangzhou, China, from May 2015 to December 2018. The study was approved by the medical ethics committee of the hospital and complied with the tenets of the Declaration of Helsinki. Written, informed consent was obtained from all patients after full explanation of the study.
Patients who had FLACS were consecutively recruited. The inclusion criteria were patients 50–90 years old with age-related cataracts. The exclusion criteria for FLACS included the presence of corneal opacity, active ocular disease, small pupils or a history of ocular trauma. Each patient had a complete preoperative ophthalmologic evaluation, including slit-lamp examinations, tonometry using a non-contact tonometer (NT-510, Nidek, Tokyo, Japan), endothelial cell density measurements with a non-contact autofocus specular microscope (EM-3000, TOMEY, Japan), corneal topography measurements with Scheimpflug imaging (Pentacam, Oculus Optikgeräte GmbH, Wetzlar, Germany), optical coherence tomography (OCT, MODEL 3000, Carl Zeiss Meditec, Jena, Germany), axial length and biometry with an A-scan ultrasound (Cinescan, Quantel Medical, Cournon d’Auvergne, France) or IOL Master (Carl Zeiss Meditec, Jena, Germany).
Surgical technique
Topical levofloxacin and pranoprofen were administered four times daily to the patients’ eyes 1 day before surgery. After pupillary dilation (one drop of 0.5% tropicamide administered three times at 15 min intervals) and topical anaesthetics (one drop of 0.5% proparacaine hydrochloride administered three times at 5 min intervals), a lid speculum was put in place and the laser was docked to the eyeball using a SoftFit patient interface (PI). The position and scope of the capsulotomy, lens prefragmentation and cornea incision were set with the guidance of the live microscope. A capsulotomy with a diameter of 5.0 mm was performed in all cases. The capsulotomy pulse energy was set at 6 μJ, with anterior and posterior offsets of 300 µm, horizontal spot spacing of 5 µm and vertical spot spacing of 4 µm. The lens nuclear prefragmentation energy was set at 10 μJ with a quadrant, sextant or grid pattern. Once the adjustments were made and the parameters were selected using the software interface, the laser treatment was initiated by pressing the footswitch.
A 2.0 mm primary corneal incision and a 1.0 mm side-port incision were made using a femtosecond laser or a keratome. Viscoelastic material was injected into the anterior chamber, and the capsulotomy was removed with a capsulorhexis forceps. In patients with incomplete capsulotomies, a secondary capsulorhexis was completed manually. After hydrodissection, the surgery was completed with standard phacoemulsification using the Stellaris system (Stellaris; Bausch & Lomb, Rochester, New York), cortex removal using automated irrigation/aspiration (I/A) and final IOL implantation in the capsular bag. Corneal incisions were hydrated with balanced salt solution for watertightness at the end of the surgery.
We designed a form to record complications during the operation. In order to avoid deviations due to memory loss, the surgeon was required to fill in this form within 5 min after each operation. Complications that needed to be recorded included suction breaks, incomplete capsulotomies (bridges and microadhesions), radial AC tears, incomplete corneal incisions (incisions that needed to be recreated with a keratome), incomplete lens prefragmentation, intraoperative miosis, posterior capsule ruptures, repeated dockings (with the number of docking attempts) and subconjunctival haemorrhages (with the number of quadrants involved). All surgery videos were kept for postoperative analysis.
Evaluation
The clinical records of the patients who underwent FLACS from May 2015 to December 2018 were collected. Detailed data on patients with incomplete capsulotomy and AC tears were extracted. Surgery videos of patients with capsulotomy-related complications were reviewed by two of the authors (WW and KY). The characteristic signs and potential causes of capsulotomy-related complications were evaluated and documented.
Statistical analysis
To characterise the learning curve, the 1600 cases of FLACS were divided into five groups in chronological order: group 1 consisted of the first 100 cases, group 2 consisted of the next 100 cases, group 3 consisted of the third group of 100 cases (ie, cases 201 to 300), group 4 consisted of the fourth group of 100 cases (ie, cases 301 to 400) and group 5 consisted of the last 1200 cases. The incidence of complications was compared between subgroups. Categorical data were tested with χ2 tests or continuity-corrected χ2 tests. Continuous data were expressed as mean±SD. The independent t test and Mann–Whitney U-test were used depending on the normality of parameters. The complication rate was evaluated using χ2 tests or continuity-corrected χ2 tests. A p value of less than 0.05 was considered statistically significant. All analyses were performed using the Statistical Package for the Social Sciences software (V.22.0, IBM Corp).
Results
In this study, we included 1600 eyes from 1140 consecutive patients who underwent FLACS performed by the surgeon (KY) between May 2015 and December 2018. The mean age of the patients was 67.28±12.92 years. Of the 1600 eyes, 52 (3.25%) had incomplete capsulotomies, of which 46 had bridges and 6 had microadhesions. Twenty-two eyes (1.38%) had AC tears, of which 12 (54.55%) occurred after an incomplete capsulotomy and a secondary capsulorhexis. AC tears extended to the posterior capsule in two eyes, in which IOLs were implanted in the sulcus after anterior vitrectomy.
Based on the video analysis, we noticed that the causes of incomplete capsulotomies could be classified into eye tilt (figure 1A)(16 eyes, 30.77%), air bubbles or ocular secretions at the interface (figure 1B) (14 eyes, 26.92%), white cataracts (figure 1C) (7 eyes, 13.46%), corneal folds (figure 1D) (5 eyes, 9.62%), laser setting errors (figure 1E) (4 eyes, 7.69%), inadequate suction (figure 1F) (3 eyes, 5.77%) and other unknown causes (3 eyes, 5.77%) (table 1).
LenSx microscopy and ocular coherence tomography (OCT) images of the anterior segment in cases prone to incomplete capsulotomy. (A) Eye tilt. The cornea and pupil are not centred in the microscopy image (left), while there is an obvious wave in the circular scan (top right) and a crystalline tilt in the linear scan (bottom right). (B) Air bubbles or secretions at the interface. Air bubbles or secretions can be identified in the microscopy image (left, white arrow), and there are usually black bands in the OCT image (blue arrow) due to light blockage. (C) A white cataract. (D) Corneal folds. Stripes in the microscopy image (left, white arrow) and discontinuity of the posterior corneal surface (bottom right, blue arrow) can be identified in the OCT image. (E) Laser setting error. The laser offset line is not parallel to the anterior capsule when the highest and lowest points of the anterior capsule setting are incorrect (top right, blue arrow). (F) Inadequate suction. The microscopy image shows a loose conjunctiva (left, white arrow) and an unclear iris at the corresponding position due to a change in focus.
A summary of the causes of incomplete capsulotomy
After being stratified in chronological order, the cases did not exhibit any statistically significant differences in patient demographics (age, sex and cataract grade) between groups (p>0.05). Throughout the study period, there were significant differences among the five groups in the incidence of incomplete capsulotomy (p<0.001) and capsule tears (p=0.0094). The incidence of incomplete capsulotomy in groups 1 and 2 was significantly higher than in group 5 (p<0.001 for group 1 vs group 5 and p=0.006 for group 2 vs group 5) (table 2), while there were no significant differences between groups 3, 4 and 5 (p>0.05). The incidence of capsule tears in group 1 was significantly higher than in group 5 (p<0.001) (table 2), but there were no significant differences between groups 2, 3, 4 and 5 (p>0.05).
Incidence of incomplete capsulotomy and anterior capsule (AC) tears across different groups of patients
Discussion
The overall incidence of incomplete capsulotomies in our study was 3.25%. Notably, the incidence in the initial 200 surgeries was 7%, and this was reduced to 2% for the subsequent surgeries. In previous assessments, where the LenSx laser was used, the rates of incomplete capsulotomies ranged from 0.76% to 20%7 9–21 (table 3). Before the introduction of the SoftFit PI and advanced software programmes, tags and bridges occurred in 20% of the surgeries.7 Although the incidence was extremely low in some studies, their small sample sizes may have underestimated the rate or neglected special cases, such as white cataracts, due to non-comprehensive patient inclusion.
Published rates of capsulotomy-related complications due to LenSx FLACS
Our study also provided the main reasons for incomplete capsulotomies and made conclusions about their characteristics based on microscopy and OCT images in the LenSx system (figure 1).
Eye tilt is recognised as one of the most common risk factors for capsulotomy errors in FLACS.9 14 A tilt of only 7.5° is sufficient to cause an incomplete capsulotomy pattern.14 OCT measurements that combine circular and linear scans provide precise information about the location of the crystalline lens. Both the obvious waves in the circular scan and the crystalline tilt in the linear scan (figure 1A) are indicative of eye tilt. Capsulotomy offsets and laser energy settings should be adjusted when these signs appear, and we suggest performing redocking once significant tilting is noticed. The extent of eye tilt-induced incomplete capsulotomy is usually large and often located in the superior area. This may be caused by Bell’s phenomenon when stressed patients try to close their eyes.
Air bubbles and ocular secretions can be clearly identified in the microscopy images . Both corneal refractive spots) and black bands in the OCT image (figure 1B) indicate the presence of air bubbles or secretions. We suggest routinely applying a drop of 0.3% sodium hyaluronate on the operated eye to wash away ocular secretions and facilitate suction before docking. It is beneficial to squeeze out the air bubbles and ocular secretions before suctioning when the front edge of the water ripple crosses the midline and forms a U shape (figure 2A). Once a closed bubble is formed between the interface and the cornea (figure 2B) before suction, the interface should be released for redocking. For small air bubbles at the interface, increasing the laser energy setting may improve the chance of a complete capsulotomy, but extra attention must be paid before and during the removal of the capsule.
Tips to prevent capsulotomy-related complications. It is beneficial to squeeze out the air bubbles and ocular secretions before suctioning when the front edge of the water ripple crosses the midline and forms a U shape (A). Air bubbles and secretions tend to block the interface after suction if a closed bubble forms (B). We injected the viscoelastic agent above the AC flap to flatten the capsule and stabilise the anterior chamber before creating the primary incision (C). Capsulorhexis outside the original incompletely cut margin (D) is recommended to avoid the formation of small tears and tags. AC, anterior capsule.
For white cataracts, especially type I mature cataracts22 with cortical liquefaction, the incidence of incomplete capsulotomy is inevitably high and does not decrease with the experience of the surgeon. In Asena et al’s study,16 small adhesions and incomplete capsulotomies occurred in as many as 16% of mature cataracts, which were much more frequent when compared with those of non-mature cataracts. Seven cases of incomplete capsulotomy in our study occurred in patients with white cataracts, 5 of which were type I mature cataracts. The immediate liberation of milky white fluid from the lens may hamper the performance of the laser, and the decreased intralenticular pressure may cause a slight change in the position of the AC plane. Both factors can influence the process of capsulotomy and result in uncut regions. Laser platform with a higher repitition rate, such as Catalys Precision system (Abbott Medical Optics Inc, Santa Ana, California),23 might be able to cut the capsule 360° before these movement occur, thus reducing capsulotomy-related risks. However, further study is required to verify this hypothesis.
Corneal folds can be identified by both stripes in the microscopy image and discontinuity of the posterior corneal surface in the OCT image (figure 1D). These folds may cause inadequate laser focusing and lead to incomplete capsulotomy, which is usually limited in extent or just microadhesion. After the introduction of the laser SoftFit PI, the occurrence of corneal folds decreased greatly when compared with a rigid curved PI.24 Apart from the variable of the PI, corneal folds are usually related to a certain degree of eyeball tilt. When the PI is suctioned from directly above, it does not generally cause corneal folds. In addition, based on our experience, patients with high myopia are more likely to develop corneal folds.
Although the setting of the highest and lowest points of the AC is usually automatically recognised by the system, errors can sometimes occur (figure 1E). This may lead to inaccurate laser focusing and, thus, incomplete capsulotomy. The surgeon should carefully observe the parameter settings and make manual adjustments if necessary.
Inadequate suction may cause eye movements during laser treatment, thus leading to incomplete capsulotomy. Potential reasons for inadequate suction include conjunctivochalasis (figure 1F), pterygium and abnormal corneal curvature. For cases with floppy conjunctiva, Nagy et al suggested pulling out the conjunctiva with a fine pincette under the patient interface.7 We found that the use of a wider eyelid speculum made it easier to effectively increase conjunctival tension and reduce conjunctival stacking.
The safety of FLACS has been questioned due to the occurrence of AC tears. High-power electron microscopy revealed postage-stamp perforations at the edge of the femtosecond laser capsulotomy,25 which was considered a risk factor for AC tears. However, with the use of SoftFit contact lenses and lower pulse energy, the condition of the capsule edges has improved26 and is approaching the smoothness of manually cut capsule edges.27 Meanwhile, the capsule edges made by the LenSx laser have proven to be of equivalent strength to manually made ones.28 There were five cases of AC tears in our initial 100 eyes (5%), but this declined to a rate of 0.83% in the final 1200 eyes, which is comparable to that of a manual capsulorhexis.29 In our study, we set the capsulotomy pulse energy at a relatively low 6 μJ in order to obtain a more regular and smoother cut surface during the creation of the capsulorhexis.30 However, the sites of the bridges and microadhesions are often the weaker areas of the capsulotomy margin that can contribute to radial tears.8 Initially, in eyes with incomplete capsulotomies, the occurrence of capsular tears after a secondary capsulorhexis was as high as 50%. With the accumulation of experience and technological progress, we reduced this incidence to 12.5%. It is very important to examine the edge of the capsulotomy to find bridges and tags in time and cope with them properly. Careful observation under higher magnification before phacoemulsification2 is recommended to confirm capsulotomy integrity. Nagy et al 7 pointed out that anterior chamber collapse during removal of the capsulotomy flap can lead to radial tears; thus, they suggested stabilising the anterior chamber with an ophthalmic viscosurgical device (OVD) before opening the main incision. In our experience, the location of the OVD injection is also important. We recommend injecting the OVD above the AC flap to flatten the capsule and avoid pushing the capsule and generating tears. This is especially important for cases with an obvious incomplete capsulotomy (figure 2C). Microadhensions are not easy to be noticed compared with bridges, and they are usually not discovered until the removal of the capsulotomy flap. Pulling towards the centre of the microadhesion may cause a tag, which can lead to an AC tear that could later extends towards the periphery.7 Therefore, pulling out the capsulotomy flap with sudden movement of a capsulorhexis forceps or phacoemulsification handpiece is not recommended, even in eyes with seemingly free floating capsule. It is safer to free the AC flap with a central dimple-down manoeuvre31 and remove it following the contour of the capsulotomy with a capsulorhexis forceps. Another point that needs to be emphasised is that surgeons should lead the capsulorhexis outside the original incompletely cut margin (figure 2D) to avoid the formation of small tears and tags,7 which may cause radial tears during later procedures. Moreover, a certain percentage of AC tears seems to be unrelated to incomplete capsulotomy. We speculate that, in these cases, microtags that cannot be recognised by the naked eye may be present, and we suggest to perform I/A a little distant from the edge of the AC, thereby avoiding aspiration of an unseen tag.32 In addition, some AC tears may be caused by improper intraoperative procedures, as that can also occur during manual surgeries. Our previous study33 on patients with white cataracts demonstrated that despite the high incidence of incomplete capsulotomy, FLACS did not result in the Argentinian flag sign and could reduce the incidence of AC tears when compared with manual capsulorhexis. For this reason, we still recommend FLACS for white cataracts. Considering the occlusion of the liberated cortex and the absence of red reflexes, we suggest membrane staining with trypan blue dye before considering a second capsulorhexis.
As a new surgical technique, FLACS involves a learning curve before attaining proficiency. Although the basics of this procedure can be mastered by initially practicing on model eyes, Christy et al recommended a training course with 25–30 real-world cases based on their study of docking time and attempts.9 In Bali et al’s study,10 the first 100 cases of FLACS performed by seven surgeons had significantly greater numbers of docking attempts and lower rates of free floating capsulotomies. Our research further proved that the FLACS capsulotomy learning curve was steep in the initial 100 surgeries, but stabilised after 200 surgeries. We confirmed that after being able to perform FLACS independently, a more advanced learning process is needed before performing stable capsulotomies.
The limitation of our study is that it involved the experience of only one surgeon; therefore, the results might be biased and less generalisable to other surgeons. In the future, studies involving multiple surgeons or multicentre prospective studies will be helpful for us to understand the FLACS learning curve and its related factors more comprehensively.
In conclusion, the main reasons for incomplete capsulotomy include eye tilt, air bubbles or ocular secretions at the interface and white cataracts. Most of the causes are predictable and preventable. Most capsulorhexis tears can be avoided if incomplete capsulotomies are recognised early and treated properly. We hope that our results and experience can help shorten the learning curve for beginners, so that ophthalmologists can master this useful technology more quickly and proficiently.
Data availability statement
Data are available upon reasonable request. Not applicable.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and was approved by the medical ethics committee of the Second Affiliated Hospital of the School of Medicine, Zhejiang University, Hangzhou, China (2014-054). Participants gave informed consent to participate in the study before taking part.
References
Footnotes
Contributors All authors conceived and designed the study. WW reviewed surgical videos, analySed data and drafted the article. XC analysed data and drafted the article. XL provided critical revision of the manuscript. XZ and DL collected data. KY performed all the FLACS, reviewed surgical videos and provided critical revision of the manuscript. All authors read and approved the final manuscript. KY is responsible for the overall content as the guarantor.
Funding This work was supported by the National Natural Science Foundation of China (81600716, 81870641, 82070939) and the Zhejiang Province Key Research and Development Program (2020C03035).
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Linked Articles
- Highlights from this issue