Introduction
Cortical removal is perhaps the most overlooked step of cataract surgery, but it is the most important step in reducing the incidence of the most common complication of cataract surgery - posterior capsule opacification (PCO). As incisions and capsulotomies have become smaller, the subincisional region has become more sequestered, making cortical removal more challenging. Whether the technique is bimanual, with separate instruments for the sideport and main incision, or a technologically advanced steerable irrigation and aspiration (I&A), the fundamental principle of cortical removal has remained the same - aspiration.
My approach to cortical removal involves a J-cannula to irrigate the cortex completely out of the capsule and usually out of the eye. In its simplest form, the irrigation will dislodge the subincisional cortex for easy aspiration from the anterior chamber. The J-cannula is capable of clearing cortex from the eye in as little as 15 seconds with only 5 mL to 10 mL of balanced salt solution. The safety of this technique is also remarkable. I have used the technique for 6 years in more than 2,500 cases without performing an anterior vitrectomy from a capsule ruptured specifically during cortical removal. (I have had one case of zonular dialysis with vitreous loss from hooking the peripheral capsule with the J-cannula. The avoidance of this type of problem is discussed later. I would posit that a 0.04% vitreous loss rate during cortical removal is acceptable.)
During hydrodissection, I use 1% preservative-free lidocaine because Luis Vargas, MD, has demonstrated that this agent decreases the adherence of cells responsible for late PCO in the rabbit model. I am careful to place the cannula tip as adjacent to the anterior capsule and as peripheral as I can, but I find I fail to completely separate the cortex from the capsule consistently. Even changing the location of my hydrodissection has not improved the consistency of my results. As an endocapsular surgeon, I am particularly disappointed when this causes the nucleus to prolapse into a semi-flip, and I must approach its removal close to the endothelium. Likewise, aggressive hydrodissection has created more than one radial tear in the anterior capsule in my hands.
Examining the "structure" of cortex allows an understanding of why hydrodissection fails. Fundamentally, cortex is a layer of organized adherent radial fibers of varying thickness. The cortical fibers display varying adherence to themselves, the capsule, and the epinucleus/nucleus. The initial wave of balanced salt solution (lidocaine) during hydrodissection acts as a wedge, with the leading edge held together by surface tension. This wave takes the path of least resistance as it traverses the space between the cortex and the capsule. Leaving the tip of the cannula, the wave passes parallel to the radially oriented fibers, and it easily strips the attachment to the capsule lengthwise. As the wave reaches the equator, this orientation remains parallel. It is at the posterior pole of the lens where the wave now finds fibers oriented perpendicular to the flow of balanced salt solution. Instead of radially stripping the cortex, the surface tension of the balanced salt solution must overcome the entire adherence of the fiber to the capsule. The path of least resistance may be between the fibers of cortex instead of between the cortex and capsule, thus sending the wave in between the cortical fibers instead of along the corticocapsular plane.
Conceptually, using a J-cannula for this technique is essentially hydrodissection of the cortex, but a bit more vigorous after the nucleus has been removed. With this technique, I regain access to the corticocapsular plane near the posterior pole of the nucleus and can pass a wave of fluid parallel to the direction of the remaining adherent fibers. This strips the cortex from its attachments, without the bulk of the nucleus to impede the flow of fluid.
Performing J-cannula irrigation is technically simple. Immediately after passing off the phaco handpiece, my staff hands me a 26-gauge McIntyre-Binkhorst J-cannula on a 5-mL or 10-mL syringe filled with balanced salt solution. I place the tip of the cannula against the capsule and irrigate. The tip is swept from side-to-side, into the capsular fornices, and the cortex is stripped from the capsule to swirl in the anterior chamber and, with practice, out of the eye. Usually, the cortex is gone from the eye in about the time my staff has prepared the automated I&A.
This technique is good for any amount of cortex, whether it is a complete cortical bowl/epinucleus stubbornly refusing to prolapse into the anterior chamber or a few whisps under the incision. If a nearly intact bowl is present, I puncture it with the tip of the cannula and may return with the phaco handpiece to remove the larger pieces. Jeff Shaver, MD, prefers to strip a small wedge of cortex opposite the incision by aspirating with the phaco tip. This simple tip provides clearer visualization to gaining an access point to the corticocapsular plane for the tip of the J-cannula.
Of particular benefit with this technique is the ability to dislodge cortical strands smaller than the aspiration port of an I&A tip. When aspirating a wedge of cortical wisps, we rely on the efficiency of the previous hydrodissection to have loosened the attachment of the cortex to the capsule. These fibers still must be greater in diameter than the aspiration port to achieve occlusion of the I&A tip and allow for efficient aspiration to occur. The irrigation again will sweep into the capsular fornices, far deeper than standard polishing techniques allow. (The J-cannulas I use regularly have a fairly aggressive sandblasted curve to allow for simultaneous polishing of the central posterior capsule while the irrigation dislodges the material from the equatorial region.)
Procedure
For the purpose of teaching, I have divided the technique into four levels of proficiency. These are basically the stages through which the J-cannula irrigation has evolved over the last few years. To avoid system errors, it is important to always lock the J-cannula onto a Luer lock syringe to prevent it from becoming a projectile. Although manufacturing defects are rare, the cannulas should be inspected initially for nicks or burrs. Have it available for immediate passing to the surgeon at the end of phacoemulsification. Most importantly, do not aspirate with this cannula at any time. With Miyake-Apple video techniques, I have demonstrated the safety of these cannulas if the capsule occludes the tip, both to the integrity of the capsule and to intact zonules. This technique is for displacing cortex, and the I&A handpiece is sufficient at removing what remains in the anterior chamber.
To Begin: Start with a cooperative patient with good dilation and a clear red reflex. Look for a region of clear capsule at about the midpoint of the pupil. Place the tip of the J-cannula under the cortex against the capsule and irrigate. The passing wave of fluid will disengage the subincisional fibers from the capsule. Even if you have not freed the cortex from its points of adhesion in the periphery, the cortex should be separated from the posterior capsule and should be much easier to grasp with standard I&A handpieces. As the cortex opposite the incision is easily reached with standard techniques, the rest can easily be cleared without further irrigation.
At this stage, it is important to be relatively gentle with the irrigation, but forceful enough to strip the cortex from the capsule. It should feel much like hydrodissection, but with different visual cues to determine the effectiveness of this step.
Second stage: The next advancement has to do with the force of the irrigation. With the syringe full and out of the eye, direct a stream of balanced salt solution across the room about 5 to 6 feet. Typically, when the anesthesiologist complains about getting wet, you have got the feel for the amount of pressure you can safely apply without risking the capsule. With the next few attempts, place the tip of the cannula as you have previously, but sweep it side-to-side against the capsule. Apply increasing pressure to the plunger of the syringe, and the subincisional cortex should start to dislodge completely. The appearance of swirling cortex in the confines of the capsular bag can be impressive.
Once this subincisional cortex is freed, it is no longer necessary to place an aspirating tip into a region of poor visualization. (And although I have heard the claims of safety made about the new silicone aspirating tips, I do not see the need to incur the risk.) It will take practice to avoid trying to rotate the I&A tip toward the posterior capsule to get that "little bit" of cortex. A second, or even third, irrigation is preferable as it will eventually strip away the cortex and prevent accidentally breaking the capsule. In a patient with a small pupil or lax zonules, this stage of the technique is invaluable for safely dislodging hidden or even adherent residual cortex.
Third stage: After becoming comfortable with completely dislodging the subincisional material, the real efficiency of the technique can be explored by dislodging all of the cortex from the adhesions of the posterior capsule. The tip of the J-cannula is directed not only toward the subincisional space, but also into the fornices of the capsule nearly opposite the incision. With practice, the entire capsular equator can be irrigated and cleared in this fashion. Dr. Shaver performs a slight variation where the IOL is now implanted and the cortex removed with the residual viscoelastic afterwards.
I had the honor of evaluating this technique with Dr. David Apple and his team led by Drs. Liliana Werner and Suresh Pandy at the Storm Eye Institute in Charleston, South Carolina. We found that this irrigation was capable of completely removing cortical debris to the level of a single residual layer of regenerative equatorial cells. The results of the study suggest that this technique may have the benefit of decreasing the rate of PCO by improving the routine efficiency of cortical removal. Although minute amounts of cortex are probably left behind on a regular basis, I am not aware of another technique that provides this level of reproducible efficiency in terms of speed and thoroughness of cleanup.
Advanced: By this point, you will undoubtedly have observed several cases where the cortex cleared itself spontaneously from the anterior chamber. Although this is fortuitous at first, the ability to do this consistently is the ultimate goal of the irrigation. As I mentioned, the swirling cortex can be impressive. By keeping the tip of the J-cannula within the confines of the capsule, the curve of the bag will also confine the swirling cortex in this space. By redirecting the flow of the irrigation slightly toward the anterior chamber and by slightly depressing the lip of the incision as you irrigate, the cortex will tend to flow out of the eye. At this point, complete cortical removal will be consistently achieved in 15 to 20 seconds after removal of the phaco handpiece.
Potential Problems
In my experience with the J-cannula, two problems have occurred on a regular, but infrequent basis. The first is hooking intraocular structures with the cannula. It is common to hook the internal incision, but this has never presented a problem. Hooking the iris and capsulotomy are much less common and tend to occur with shallowing of the chamber and/or depletion of the irrigating fluid. Rarely, I will have to reform the anterior chamber with balanced salt solution or viscoelastic from the sideport to free the J-cannula. I have hooked the peripheral capsule on two occasions, with one resulting in zonular dialysis and vitreous loss. Both were awkward superior approaches, with the patient achieving exaggerated gaze first in the direction of the incision to place the cannula in the wrong location, then moving the eye quickly to inferior gaze. In the case of zonular dialysis, the patient continued redirecting her gaze before I could respond. In the other, I was able to gently unhook the capsule without consequence.
As one becomes more aggressive with the technique, iris prolapse can become an issue. This is now an infrequent problem since moving to a limbus-based incision and topical/subconjunctival anesthesia some time ago. (My incision over the last few years has been a 2.7-mm to 2.8-mm limbus-based hinge with a 2.0-mm to 2.5-mm tunnel length.) These changes have improved my anatomic entry away from the base of the iris and significantly reduced posterior pressure. The only other factor I can isolate is failure to use phenylepherine for dilation. (An anesthesiologist convinced me it would help my patient's blood pressure if I did not use phenylepherine for dilation for 1 day's surgery. However, I believe it raised mine.) Iris prolapse still occurs and I will attempt to dislodge the subincisional cortex with the J-cannula. In these cases, I am gentle with the irrigation. I will typically follow up with standard aspiration techniques or even IOL rotation to clear the cortex. My best advice is to be much more cautious with the irrigation in eyes with less-than-wonderful dilation and to watch the incision carefully, as ceasing to irrigate at the first sign of abnormal iris position will prevent any functional or cosmetic problems.
Surprisingly, the irrigation can be vigorous and still remain safe to a damaged capsule. I have videotaped cases in which single occult radial tears were present in the capsulorrhexis. I have irrigated vigorously, even directly in line with the tear, and have yet to see a tear extend due to the irrigation alone. This finding surprised me, but it goes hand-in-hand with my observed improved safety of this technique. Obviously, if I am aware of a radial tear or if I use a can-opener capsulotomy, I significantly decrease the force of my irrigation.
Overall Benefits
It is difficult for me to decide which is the best motivating factor for the use of this technique: speed or safety. I have always tried to make both points when discussing the irrigation, including the minimal amount of balanced salt solution needed to remove the cortex. The J-cannula is about the most inexpensive item in my surgical tray and it fits neatly into any surgical regimen to stage any method of cortical removal. The ability to remove cortex consistently without direct visualization has been a significant benefit, especially in challenging cases.
Posted September 2003