14 Optical Coherence Tomography Analysis in Cataract Surgery
Optical coherence tomography (OCT) has extensive applications in the posterior segment (PS), and development of this technology for the anterior segment (AS) also has particular uses in cataract surgery. 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 It is interesting to note, however, that in a review article from 2011 on AS OCT, cataract surgery was not mentioned at all. Initially, OCT had been used to visualize the whole AS; then it was found that corneal wounds could be studied. More recently, with the use of the femtosecond laser for cataract surgery, OCT has found a vital role in imaging the AS to set the boundaries of laser activity in performing incisions, capsulotomy, and nuclear segmenting. Before we look at these in greater detail, remember the role of OCT retinal imaging to detect macular problems before cataract surgery. This is particularly important where multifocal intraocular lens (IOL) implantation is being considered. After cataract surgery, such macular changes as cystoid macular edema can be evaluated and the clinical course followed. AS OCT has also been used to study iris architecture, posterior capsular opacity, and anatomical changes in relation to specific postoperative issues.
14.1 Measuring the Anterior Segment
Anterior-segment OCT can be used to measure a number of different parts of the AS (Fig. 14.1), which can be useful in cataract surgery as follows:
Although the depth of the anterior chamber is generally measured as part of the biometry process, whether optically or with ultrasound, there is a place for OCT also. In a recent study, however, significant differences were found between the actual measurements recorded by different means. Partial coherence inferometry and Scheimpflug measurements were comparable, as were both manual and automated AS OCT but not each pair with the other.
In addition to the depth of the anterior chamber, the curvature of the cornea, and thus its refractive power can be measured using AS OCT by matching curves to the anterior and posterior surfaces of the cornea. It has been found particularly useful for IOL calculations in patients who have had previous myopic laser refractive surgery. In a study comparing the use of AS OCT corneal measurements to obtain true corneal power with the Haigis-L formula (specifically for postmyopic laser patients), contact lens over refraction, and clinical history methods for the IOL calculation, the AS OCT results were much more accurate. The mean absolute error was 0.5 for AS OCT, 0.78 for Haigis-L, 1.46 for contact lens over refraction, and 1.78 for the clinical history method.
Although toric IOLs are becoming more commonly used, keratotomies are still frequently used and ASOCT can be used for pachymetry of corneal thickness. This should help to prevent unexpected and certainly unwanted perforations. However, a difference was found when toric IOLs were compared with ultrasound measurements, and they cannot be used interchangeably because the AS OCT measured significantly less mean CCT of 536.9 ± 27.0 µm for OCT and 556.6 ± 30.5 µm for ultrasound.
Lens thickness can also be measured with AS OCT, which is useful for hypermature cataracts before surgery to make the surgeon aware that the capsule may be under tension and the necessary precautions to avoid an anterior capsular tear can be implemented. With the advent of femtosecond laser–assisted cataract surgery (FLACS), the size of the lens is also vital. Most of the currently available lasers use AS OCT to visualize the lens in cross-section to set the extent of laser segmentation of the nucleus.
14.2 Observing Anatomical Changes in the Anterior Segment Caused by Cataract Surgery
A number of investigators have looked at the size, shape, and behavior of anterior chamber structures before and after cataract surgery using ASOCT:
The appearance of the anterior chamber angle before and after phacoemulsification has been studied and showed, across a range patients aged between 15 and 91 years, that the angle-opening distance and the iris-trabecular space increased in all of them (Fig. 14.2). Further, the amount of change was greater in older patients. The authors postulated that the greater lens volume with age might account for this finding.
In another study to try to predict postoperative intraocular pressure (IOP), the iris cross-sectional area and convexity as seen with ASOCT proved useful predictors. A high cross-sectional area or convex hull of the iris segments was associated with lower postoperative IOP.
It has been postulated that the timing and formation of the capsular bend around the sharp edge of an IOL may be important in preventing posterior capsular opacity. AS OCT has been used to study capsular bag closure in the early postoperative period; a one-piece hydrophobic IOL had earlier capsule bend formation at the optic edge than a three-piece silicone IOL.
Late-onset capsular block syndrome occurs when milky-white liquid accumulates behind the IOL and reduces the patient’s vision. The appearances before and after uneventful neodymium:yttrium-aluminum-garnet (Nd:YAG) laser capsulotomies were observed using AS OCT (Fig. 14.3). Although the patient’s best-corrected visual acuity (BCVA) improved in all cases, the IOL position and refraction did not change.
The thickness of the posterior capsule with different IOL designs and materials has also been studied. A round-edged polymethylmethacrylate (PMMA) IOL was found to have the thickest posterior capsule, followed by a hydrophilic lens; the thinnest was the hydrophobic IOL. This correlated with BCVA.
More recently, with the use of the FLACS, OCT has found a vital role in AS imaging to set the boundaries of laser activity in performing incisions, capsulotomy, and nuclear segmenting. Considerable discussion has appeared in the peer-reviewed literature about the way different clear corneal incision (CCI) architectures behave during and after cataract surgery, and AS OCT has given us the tool to investigate this topic. Many aspects of the incision have now been studied from cross-sectional shape, dimensions, consistency, distortability, damage to corneal structures, ability to seal, effects of stromal hydration, long-term healing, comparison between laser and knife incisions, and many others. This part of the chapter presents an overview of the main findings.
14.3 Anterior-Segment OCT and the Cataract Incision
The first OCT study of corneal wounds in cataract surgery was done in 2006 using a modified retinal OCT; however, it demonstrated some of the possibilities in terms of imaging the cross-sectional architecture. The wounds were evaluated at 1, 3, and 30 days after surgery, and the most striking finding was that even though poor endothelial apposition was common, it did not lead to wound leakage even in these 3.2-mm incisions.
The following year, a study was reported using the first dedicated AS OCT, the Zeiss Visante. The object of this study was to study the profile of the wound created with different knives. Although the incisions were supposed to be uniplanar, all showed an arcuate configuration (Fig. 14.4), described as being like a “tongue and groove” in the cornea that helped the incision to close. Further, with this arcuate profile, there was an extended arc of contact over a straight one.
This study had been performed 1 day after cataract surgery. Given that it is postulated that the early hours after surgery may be the time of greatest vulnerability for the wound, a study of early wound behavior seemed important. Patients were examined 1 hour after cataract surgery. The Visante was again used, and the essential features of the post-cataract surgical wound were defined. Five CCI architectural features were noted with the following frequencies: epithelial gaping (12%), endothelial gaping (41%), endothelial misalignment (65%), local detachment of Descemet membrane (DM) (62%), and loss of coaptation (9%) (Fig. 14.5). This study was across four different wound sizes, from 3.2 to 2.2 mm. Interestingly, when IOP was measured at 1 hour, ranging from 3 to 46 mm Hg, the architectural features were related to IOP, not to wound width.
Multiple studies have now shown the repeatability of the dimensions of corneal wounds, such as chord length, incision angle, and corneal thickness at the incision (Fig. 14.6). These findings, combined with observation of the architectural features mentioned already, have become the standards used for any AS OCT study of cataract incisions.
Several studies have now compared wounds in standard small-incision phacoemulsification using coaxial microincision cataract surgery with wounds using biaxial microincision cataract surgery. There seems to be some disagreement as to whether the biaxial or coaxial incisions cause more disturbances to the corneal architecture both in the short term or longer term, but none of these differences reach statistical significance.
As a further protection for the CCI in the early postoperative period, hydrogel materials have been developed to seal the wound. These are called adherent ocular bandages (AOBs), and AS OCT has been used to assess their behavior. In one study, patients who underwent coaxial microincision cataract surgery were allocated to an AOB group or to a control group. The CCIs were examined postoperatively within 2 hours, at 24 hours, and at 7 days using OCT imaging and a slit-lamp fluorescein 2% Seidel test. A significant difference was found in the mean immediate postoperative IOP compared with the control group (13.4 mm Hg ± 5.28 [SD]; range, 5–23 mm Hg) and the bandage group (19.4 ± 5.94 mm Hg; range, 11–29 mm Hg) (P <0.001, t test). This finding may mean that the AOB was preventing microleaks (Fig. 14.7). It was further found that the AOBs protected the incisions, selectively adhering to de-epithelialized areas and rapidly clearing from re-epithelialized areas.
So far, we have looked at incision architecture in single slices, which is useful for seeing the profile of a corneal wound; however, some machines now have a raster program that enables a block of tissue to be analyzed as a series of narrow slices right across the wound (Fig. 14.8), providing the possibility to see whether the architecture that is normally observed at the midpoint of the wound is different at the edges. Indeed, this has proven to be so with three plane wounds becoming two planes at the edges. Also, if there is to be loss of coaptation, it is more likely to be found at the edges. This sort of study should be useful in designing better knives for corneal incisions. We carried out a study using AS OCT to analyze a consecutive cohort of 30 2.2-mm CCIs made with a new knife with novel blade configuration. It was possible to show that 29 of 30 were three plane incisions and the variability of wound length was only ± 0.12 mm.
In addition, AS OCT has been used to assess corneal wound healing in the longer term. Images of wounds were obtained in consecutive eyes that had phacoemulsification 1 day to 180 months previously. The presence of DM detachment, posterior wound gape, and posterior wound retraction was assessed. The depth of wound retraction along the incision and the radial length of the incision were measured. The percentage of wound retraction relative to radial incision length was calculated. DM detachment and posterior wound gape appeared in the early postoperative period and persisted for up to 3 months, whereas posterior wound retraction developed later and was present in more than 90% of eyes after 3 years, indicating long-term wound remodeling (Fig. 14.9).
The effect on a CCI by IOL injection has also been investigated using AS OCT in a study to analyze the effects of speed in lens insertion. Eyes that had phacoemulsification and Acrysof IQ IOL implantation using a screw-plunger type injector were randomly divided into two equally sized groups as follows: group F, fast IOL insertion (one revolution per second [rps]) plunger speed and group S, slow IOL insertion (¼ rps). When an injector system was used, slow IOL insertion affected clear corneal wound structure more than fast IOL insertion did, and twice as many eyes required stromal hydration (21 of 40 compared with 11 of 40). This was statistically significant (P = 0.003). 9
Use of AS OCT with a raster program can assess wound size, and thus wound stretch, with great accuracy. In an unpublished study presented at the American Society of Cataract and Refractive Surgery in 2010, we examined at wound size 2 hours after phacoemulsification using the RTVue OCT. Seventeen slices at 15-µm intervals demonstrated in a 2.5-mm cube that the 2.2-mm incision was at that size and that any stretch from the surgery and lens implantation had gone. Interestingly, all the lenses were inserted with a wound-assisted technique using a one-handed injector and countertraction, which is appreciably faster than can be achieved with a screw injector.
The femtosecond laser, as we have seen, is now used to assist in cataract surgery, and AS OCT is an important part of imaging the AS in the process; but AS OCT has also been used to assess and compare the wounds created by this laser technology with incisions made with a steel blade (Fig. 14.10).
It has been reported that femtosecond laser–generated CCIs had significantly lower endothelial gaping, endothelial misalignment, DM detachment, and posterior wound retraction than keratome-created CCIs and were within 10% of the intended length, depth, and angle measurements. 8 As far as outcomes were concerned, however, no difference was seen. Another study using cadaver eyes to test the sealability of CCIs made with either the laser or manually showed no difference in sealability despite the fact that the laser incisions were closer to target geometry and showed less variability than the manual ones.