February 2017

 

RETINA

 

Retina consultation corner

Retinal evaluation before cataract surgery


by Steve Charles, MD

 
 

Steve Charles, MD
Steve Charles, MD

 
Optic nerve head
Optic nerve head and nerve fiber layer analysis of a glaucomatous patient
Source: Mohammad Rafieetary, OD, FAAO

Catching visual surprises before they thwart high quality cataract surgery

Currently there is substantial and appropriate emphasis on refractive surprises after cataract surgery. Advanced corneal optical properties and axial length measurement technology, more precise algorithms, an artificial intelligence engine combining data from many sources and multiple algorithms, 3-D OCT combined with femtosecond laser, guidance systems for toric IOL rotational alignment, and intraoperative aberrometry have all contributed to more predictable refractive outcomes. Patient satisfaction is driven in large part by optimal refractive outcomes.
Not uncommonly, high quality cataract surgery with a near perfect refractive outcome produces disappointing visual function because of pre-existing retinal pathology not discovered before cataract surgery. I use the term visual surprises for this unfortunate situation. There are many causes of this relatively common problem. Visually significant cataract obviously greatly impacts retinal visualization. Co-management can play a role if the referring optometrist or non-surgical ophthalmologist is unable to perform a sufficient retinal examination. There are so many techniques and so much technology for today’s refractive cataract surgeon to master it is not surprising that retinal examination may not be a core competency. Wide-field retinal imaging technology is not a substitute for careful peripheral retinal examination with a widely dilated pupil and indirect ophthalmoscope.
Cataract surgery patients are rarely told that 0.5 to 1.0% of cataract surgery patients develop retinal detachments. Many post-cataract surgery retinal detachments are preventable by preoperative retinal examination and laser or cryo prophylaxis of retinal breaks. A large prospective of retinal examination before LASIK project executed by the Pan-American Collaborative Retina Study (PACORES) Group discovered substantial peripheral retinal breaks, which were treated with laser retinopexy. In a sense, LASIK reduced the natural history of myopic retinal detachment because it resulted in pre-LASIK retinal examination and retinal break prophylaxis that otherwise might not have been done. Retinal breaks and retinal detachments are not caused by longer axial lengths; axial length is a proxy for lattice degeneration that may or may not be present. The conventional teaching that only symptomatic flap (horseshoe) tears should be treated with retinopexy is obsolete. Laser retinopexy is incredibly safe and effective in preventing retinal detachment; I do not recommend use of a retrobulbar block. Laser indirect ophthalmoscope (LIO) is usually easier to use than slit lamp delivery in cataract patients. There is no excuse for not examining the periphery and treating retinal breaks with laser or referring the patient to a retinal specialist if there is any question.
Another crucial driver of visual surprises after surgery is the prevalence of macular pathology that is invisible with the slit lamp and 90 D lens or the indirect ophthalmoscope. Fluorescein angiography is rarely indicated since OCT was introduced. OCT routinely reveals significant macular pathology invisible by ophthalmoscopy, and it is amazing what can be imaged through dense cataract, dense asteroid hyalosis, and even moderate vitreous hemorrhage. Time domain OCT is obsolete; use of spectral domain or swept source OCT is mandatory. I think all patients should have OCT before cataract surgery, not just premium IOL patients. Pseudo-color displays hide information and produce artifacts; gray scale display, white images on a black background, is far better than pseudo-color. 3-D rendering is misleading and should never be used. Technicians should never select an image to incorporate in the patient record; the ophthalmologist should view all B-scan OCT slices using the native OCT imaging software. Thickness maps that resemble corneal topography are not appropriate for clinical use because many diseases cause increased thickness, not just edema. Retinal specialists using spectral domain OCT often recognize otherwise invisible vitreomacular traction syndrome, vitreomacular schisis, transparent epimacular membranes, subretinal fluid caused by CNV from wet AMD as well as other disease processes, and increasingly common late onset central serous chorioretinopathy, probably due to overuse of steroids delivered by many routes. Discovery of these otherwise invisible disease processes can result in early intervention, which is crucial with wet AMD. The presence of medium to large drusen visible on OCT is a contraindication for the use of multifocal IOLs. Use of multifocal IOLs even with small or intermediate drusen especially in younger patients is a relative contraindication because drusen typically progress with each decade of life; geographic atrophy and wet AMD may appear in upcoming years if not present at the time of cataract surgery.
More precise optical axial length measurements that measure from the RPE instead of A-scan ultrasound, which measures from the elevated retinal surface, are crucial if macular disease is present on OCT. I think that phaco-vit is over utilized; I recommend cataract surgery before vitrectomy if there is 3+ NS or any PSC with the proviso that axial length should be measured optically. If there is 1–2+ NS, I recommend vitrectomy before cataract surgery.
In summary, using spectral domain OCT as an essential element of pre-cataract surgery evaluation will improve the quality of care and patient satisfaction.

Editors’ note: Dr. Charles is chair of the ASCRS Retina Clinical Committee. He has financial interests with Alcon (Fort Worth, Texas).

Contact information

Charles: scharles@att.net

Retinal evaluation Retinal evaluation
Ophthalmology News - EyeWorld Magazine
283 110
220 170
,
2017-04-18T04:18:22Z
True, 2