Management of Astigmatism in Conjunction with Lens Surgery
LOUIS D. NICHAMIN
Table Of Contents
PATIENT SELECTION AND CONSIDERATIONS
LIMBAL RELAXING INCISIONS
|When assessing recent changes in modern cataract and intraocular lens implant (IOL) surgery, arguably the single most pressing challenge
facing today's phacoemulsification (phaco) surgeon
is the need to achieve predictable and accurate refractive outcomes. Surgeons
and patients alike have come to largely measure the success
of their surgery by the refractive outcome, and one of the leading causes
for litigation in this field is the “refractive surprise.”1 In addition, the refractive lens exchange has become an important component
of the refractive surgeon's armamentarium.2 As such, the fields of cataract and refractive surgery have merged to
form an amalgam without distinct borders. Improved refractive results
have come about by way of improvements in both surgical technique and
advances in technology. Spherical results, for example, have become more
predictable because of increased attention to biometry technique, as
well as breakthrough technology such as partial coherence interferometry.3|
No less important is the astigmatic component of the refractive equation. At one time during the evolution of small incision surgery, it was the surgeon's goal to not induce astigmatism.4 Today, in order to fully embrace the concept of “refractive cataract surgery,” one must be able to address and reduce significant preexisting cylinder.
|PATIENT SELECTION AND CONSIDERATIONS|
|Estimates of the incidence of significant, naturally-occurring astigmatism
vary widely from 7.5% to 75%.5 It is thought that 3% to 15% of eyes may have two or more
diopters (D) of astigmatism.6 In light of recent experience gained in the field of refractive surgery, many
surgeons would agree that astigmatism of greater than 0.5 D will
lead to symptoms of ghosting and shadows. Although the older cataract
patient may be more tolerant of cylinder, the ambitious refractive
cataract surgeon should likely approach an implant patient with the same
high goals that he or she might with a younger keratorefractive candidate. Indeed, successful
cataract practices are now aiming for both
spherical and astigmatic outcomes ± 0.5 D.7|
When considering astigmatism correction, one must take into account the location of the cylinder, age of the patient, and status of the fellow eye. Because most patients will drift against-the-rule (ATR) over their lifetime—for example, toward plus cylinder at 180 degrees—many surgeons advocate a less aggressive approach to the reduction of with-the-rule (WTR) cylinder. Some authors have suggested that residual WTR astigmatism may favor better uncorrected distance acuity, given that most visual stimuli are of a vertical nature.8 Similarly, it has been contended that ATR cylinder may improve uncorrected near vision.9 The tenet that residual (myopic) WTR astigmatism is a desirable goal in order to enlarge the conoid of Sturm and hence optimize depth perception has, however, recently been called into question .10 Currently, with recent refinements in surgical technique, a spherical goal may be most desirable for the majority of patients undergoing implant surgery.
|The first decision faced by the surgeon is whether to address preexisting
astigmatism at the time of cataract and IOL surgery, or to defer and
treat the cylinder separately. One could reasonably argue that for optimal
accuracy, sufficient time for wound healing should be allotted
and a stable refraction ought to be documented prior to astigmatic correction. This
consideration is more germane with the use of rigid implants
and larger incisions. Currently, most surgeons are utilizing foldable
IOLs and studies have well documented the nearly neutral astigmatic
effect that these incisions bear when kept at or near 3.0 mm, as well
as their early refractive stability.11,12 As such, many surgeons feel that concomitant treatment of preexisting
astigmatism is a more efficient approach and is favored because it will
likely save the patient from having to undergo a second procedure.|
The next major decision is whether to treat the astigmatism through a lenticular approach (that is, to employ a toric IOL) or to utilize a keratorefractive technique. From a theoretical perspective, it is hard to argue against the use of a toric implant and their effectiveness has been widely reported.13 This option has the potential to avoid induced irregular astigmatism from corneal manipulation and provides the option of reversibility. In the United States, only one toric IOL has been approved, a single-piece plate haptic design manufactured by Staar Surgical (Fig. 1). The implant is available in two toric powers of 2.0 and 3.5 D. Propitious outcomes have been obtained with this device even with minimal experience by community-based surgeons.14 For surgeons using this particular toric implant, lens rotation is a recognized problem; Sun and co-workers reported a need to return to the operating room for repositioning in 9.2% of cases.13 Ruhswurm further reported axis rotation of at least 25 degrees in 18.9% in their series.15 According to Euler's theorem, an axis deviation of 5, 10, or 15 degrees will result in respective 17%, 33%, and 50% reductions, in effect.5 Optimal timing of the IOL repositioning would appear to be between 1 and 2 weeks following implantation as capsular fibrosis is underway, and may serve to permanently fixate the toric device in the proper meridian. Additionally, some surgeons have avoided this particular implant because of its plate haptic design and for the first-generation silicone elastomer from which it is comprised. Additional designs are available internationally and are also expected to be approved soon by the Food and Drug Administration.16 These newer designs may offer better rotational stability and therefore may see increased use.
The notion of reducing astigmatism by way of adjunctive keratorefractive surgery, specifically astigmatic keratotomy, dates back to the mid-1980s.17–19 Throughout the 1990s, a number of authors recognized the advantages of moving corneal astigmatic relaxing incisions peripherally toward the limbus.20–22 These so-called limbal relaxing incisions (LRIs) have become the most popular way to manage astigmatism at the time of cataract surgery and will be addressed in detail below.
Another viable option to decrease astigmatism is to manipulate the cataract incision by first placing it upon the steep corneal meridian, and then by varying its size and design, affect a desired amount of wound flattening, and hence a decrease in cylinder.23 Specifically, one can increase or decrease the amount of wound flattening by increasing or decreasing the size of the incision. Similarly, wound flattening may be enhanced by moving closer to the visual axis, or by creating a more circum-parallel incision to the limbus. Also, a perpendicular component, or groove, may be added to the incision to further increase wound flattening and “against-the- wound” astigmatic drift (Fig. 2).24 This approach, however, presents logistical challenges including movement around the surgical table, often producing awkward hand positions. In addition, varying surgical instrumentation may be needed along with a dynamic mindset. For these reasons, this technique has largely been supplanted by the use of a consistent and essentially neutral phaco incision, typically located temporally for astigmatic stability, and then adding supplemental relaxing incisions (LRIs).
Several other options to reduce astigmatism deserve mention. Lever and Dahan have suggested the novel use of opposing clear corneal incisions to address preexisting cylinders.25 In this technique, a second opposite penetrating clear corneal incision is placed over the steep meridian, 180 degrees away from the main incision. This approach is technically simple and requires no additional instrumentation; however, a second substantial penetrating incision is now present, possibly increasing the risk of wound leak or even infection. In addition, single-plane beveled incisions are known not to be as effective for a given arc length at flattening the cornea as more conventional perpendicular relaxing incisions.24,26 Another recently described alternative, “bioptics,” is gradually gaining in popularity, particularly with refractive lens exchange patients. With this technique, one intentionally takes a staged approach using excimer laser technology to reduce both residual astigmatic as well as spherical refractive error.27,28 Finally, conductive keratoplasty used in an off-label fashion has also recently been described as a means by which hyperopic astigmatism may be effectively reduced.29
|LIMBAL RELAXING INCISIONS|
|The first description of the astigmatic effect of nonpenetrating incisions
placed near the limbus dates back to 1898 and is credited to the Dutch
ophthalmologist L.J. Lans.30 As noted, LRIs have become the most popular technique employed today to
reduce pre-existing astigmatism at the time of implant surgery. Although
our preference is to use a temporal single-plane clear-corneal phaco
incision, one may utilize LRIs with any type of cataract incision
as long as the astigmatic effect is known and factored into the surgical
plan. LRIs offer several advantages over astigmatic incisions placed
within the cornea at smaller optical zones. These would include less
chance of causing a shift in the resultant cylinder axis. This presumably
is due to a diminished need for precise centration upon the steep
meridian. More importantly, there is less of a tendency to cause irregular
corneal flattening, and hence less chance of inducing irregular
astigmatism. Technically, LRIs are easier to perform and more forgiving
than shorter and more central corneal astigmatic incisions, and patients
generally report less discomfort. Another important advantage gained
by moving out to the limbus involves the “coupling ratio,” which
describes the amount of flattening that occurs in the incised
meridian relative to the amount of steepening that results 90 degrees
away. It has been our experience that paired LRIs (when kept at
or under 90 degrees of arc length) exhibit a very consistent 1:1 ratio, and
therefore elicit little change in spheroequivalence, obviating
the need to make any change in implant power.|
Admittedly, these more peripheral incisions are less powerful, but are still capable of correcting up to 3 to 4 diopters of astigmatism in the cataract-age population. One must keep in mind that the goal is to reduce the patient's cylinder, without overcorrecting or shifting the resultant axis. To achieve a given amount of correction, these peripheral intralimbal incisions must be longer in total arc length than more centrally-placed corneal astigmatic incisions; however, unlike longer radial keratotomy incisions, they appear to be stable with regard to refractive effect and show little sign of inducing problems such as dry-eye syndrome or other pejorative effects from corneal denervation.22 Their stability may be due to the proximity of well-vascularized limbal tissue. There are, of course, potential complications with any surgical technique and these are addressed below.
Perhaps the most challenging aspect of astigmatism surgery involves the determination of the quantity and exact location of the cylinder that is to be corrected, and thereby formulating a surgical plan. Unfortunately, preoperative measurements—keratometry, refraction, and corneal topography—do not always correlate. Lenticular astigmatism may account for some of this disparity, particularly in cases where there is a wide variance between refraction and corneal measurements; however, some discrepancies are likely due to the inherent shortcomings of traditional measurements of astigmatism. Standard keratometry, for example, measures only two points in each meridian at a single optical zone of approximately 3 mm.
When confounding measurements do arise, one may compromise and average the disparate readings; for example, if refraction shows 2 D of astigmatism and keratometry reveals only 1 D, it would be reasonable to correct for 1.5 D. Alternatively, if preoperative calculations vary widely, one may defer placing the relaxing incisions until a stable refraction postimplantation is obtained, and then correct any remaining astigmatism as needed. LRIs may be safely performed in the office in an appropriate treatment-room setting. Corneal topography can be very helpful when refraction and keratometry do not agree, and it is increasingly becoming the overall guiding measurement upon which the surgical plan is based. Topography is also helpful in detecting subtle corneal pathology such as keratoconus fruste, which would likely negate the use of LRIs, or subtle irregular astigmatism such as that caused by epithelial basement membrane dystrophy.
Once the amount of astigmatism to be corrected has been determined, a nomogram must be consulted to determine the appropriate arc length of the incisions. A number of popular nomograms are currently available.31 Our nomogram of choice originated from the work of Dr. Stephen Hollis and incorporates concepts taught by Spencer Thornton, M.D., particularly his age modifiers.24 As seen in the nomogram, a patient is considered to be “spherical” if they have up to 0.75 D of with-the-rule or 0.50 D of against-the-rule astigmatism, in which case a single plane temporal clear corneal incision is used without additional wound manipulation (Table 1). If the patient has more than this amount of cylinder, one determines whether it is WTR (45 to 135 degrees) or ATR (0 to 44 or 136 to 180 degrees) and then consults the appropriate section of the nomogram. One aligns the patient's age with the amount of preoperative cylinder to be corrected and finds the suggested arc length that the incisions should subtend.
Empiric blade-depth setting is 600 microns. When placing intralimbal relaxing incisions following or concomitant with radial relaxing incisions, total arc length is decreased by 50%.
*Nasal limbal arc only.
aUp to +0.75 × 90 or +0.50 × 180.
bSteep Axis 0 to 44 degrees or 136 to 180 degrees.
cSteep Axis 45 to 135 degrees.
All incisions are paired, except in the case of very low ATR astigmatism wherein a single 35-degree nasal LRI is placed opposite to the single-plane temporal clear-corneal phaco incision. Paired incisions are preferred to optimize symmetric corneal flattening and are expressed in degrees of arc rather than chord length. This is done in order to diminish over- and undercorrections for unusually small or large corneas, because corneal diameter may significantly impact the relative length of the arcuate incision and its resultant effect (Fig. 3). This nomogram, which has been designed specifically for the cataract patient, is based upon the use of an empiric blade depth setting of 600 microns. Individual surgeon technique and blade style may impact results, and thereby require adjustment of the nomogram. A second, slightly more aggressive nomogram is used with younger patients, particularly in the setting of refractive lens exchange surgery or in conjunction with LASIK for the correction of higher levels of astigmatism (Table 2). In this setting where optimal precision is mandated, pachymetry is performed over the entire arc length of the intended incision site, and a diamond blade with an adjustable micrometer is set to 90% of the thinnest reading obtained. The NAPA nomogram, pachymetry, and adjusted blade depth settings may certainly be used with the cataract patient, but the small compromise that is made in using an empiric blade depth setting is felt to be acceptable in this older patient population in light of increased OR efficiency.
When placing intralimbal relaxing incisions following or concomitant with radial relaxing incisions, total arc length is decreased by 50%.
aSteep axis 45 to 135 degrees.
bSteep axis 0 to 44 degrees or 136 to 180 degrees.
Some surgeons prefer to perform LRIs at the conclusion of surgery in the event that a complication occurs necessitating a modification to the phaco incision. For routine cases, however, our preference is to place these relaxing incisions at the outset of surgery in order to minimize epithelial disruption. The one exception to this rule occurs in the case of high ATR astigmatism wherein the nomogram calls for a temporal arcuate incision greater than 40 degrees of arc. Because the temporal arc will be superimposed upon the phaco incision, if it is extended to its full arc length at the start of surgery, significant gaping and edema may result secondary to intraoperative wound manipulation. In this setting, the temporal incision is first made by creating a two-plane grooved phaco incision (at 600 micron depth). Following IOL implantation and prior to viscoelastic removal, while the globe is still firm, the relaxing incision is extended to its full arc length as dictated by the nomogram. When an LRI is superimposed upon the phaco tunnel, the keratome entry is first accomplished by pressing the bottom surface of the keratome blade downward upon the outer or posterior edge of the LRI. The keratome is then advanced into the LRI at an iris-parallel plane. This angulation will promote a dissection that takes place at midstromal depth, which will help assure adequate tunnel length and a self-sealing closure.
Proper centration of the incisions over the steep corneal meridian is of utmost importance. Increasing evidence supports the notion that significant cyclotorsion may occur when assuming a supine position.32 As previously noted, an axis deviation of only 15 degrees may result in a 50% reduction of surgical effect.5 For this reason, most surgeons advocate placing an orientation mark at the 12:00 or 6:00 limbus while the patient is in an upright position. This is particularly important when employing injection anesthesia wherein unpredictable ocular rotation may occur. An additional measure that may be employed to help center the relaxing incisions is to identify the steep meridian (plus cylinder axis) intraoperatively using some form of keratoscopy. The steep meridian over which the incisions are to be placed corresponds to the shorter axis of the reflected corneal mire. A simple handheld device such as the Maloney (Storz, Katena) or Nichamin (Mastel Precision) keratoscope works well, or a more robust and well-defined mire may be obtained through an elaborate microscope-mounted instrument such as the Mastel Ring of Light (Mastel Precision). Another common way in which the steep meridian is marked utilizes a Mendez Ring or similar degree gauge that is aligned with the previously placed limbal orientation mark, and the cylinder axis is then located on the 360-degree gauge.
The LRI should be placed at the most peripheral extent of clear corneal tissue, just inside of the true surgical limbus. This holds true irrespective of the presence of pannus or blood vessels. If bleeding occurs, it may be ignored and will cease spontaneously. One must avoid placing the incisions further out at the true surgical limbus in that a significant reduction of effect will likely occur due to both increased tissue thickness and a variation in tissue composition; these incisions are, therefore, really intralimbal in nature. In creating the incision, it is important to hold the knife perpendicular to the corneal surface in order to achieve consistent depth and effect. Good hand and wrist support is important, and the blade ought to be held as if one were throwing a dart such that the instrument may be rotated between thumb and index finger as it is being advanced, thus leading to smooth arcuate incisions. Typically, the right hand is used to create incisions on the right side of the globe, and the left hand for incisions on the left side. In most cases, it is more efficient to pull the blade toward oneself, as opposed to pushing it away. A lightly moistened corneal surface will help to prevent epithelial disruption, but may mask an unintentional perforation.
The extent of arc to be incised may be demarcated in several different ways. Our preferred method makes use of a modified Fine-Thornton fixation ring (Nichamin Fixation Ring and Gauge; Mastel Precision, Storz, Rhein Medical). This instrument serves to fixate and position the globe in order to optimize incision placement, as well as to delineate the extent of arc to be incised. One visually extrapolates from the limbus to marks on the surface of the ring. Each incremental mark is 10 degrees apart, and bold hash marks (180 degrees) opposite to each other serve to align and center the incision over the steep meridian. This approach obviates the need to ink and physically mark the cornea. If one desires, particularly when first gaining experience with LRIs, a two-cut RK marker may be used to place ink marks upon the cornea to show the exact extent of arc that is to be incised, in conjunction with the fixation ring/gauge (Fig. 4). Alternatively, various press-on markers are available, such as those made by Rhein Medical (Dell-Nichamin Marker, Nichamin-Kershner Marker, or the Ruminson Marker) (Fig. 5). ASICO and other instrument companies offer a full line of dedicated markers, rings, and blades for performing LRIs.
As noted, in the setting of concomitant cataract surgery, an empiric blade depth setting of 600 microns is commonly employed. Various knives have been designed specifically for this application, ranging from disposable steel blades to exquisite gemstone diamond knives. Synthetic (and less expensive) diamond materials are also available and are intended for limited reuse. Our preference is for diamond blade technology that incorporates a single small and arced footplate for enhanced visualization at the limbus (Mastel Precision). Two models are available, one with a preset depth of 600 microns and the other with an adjustable micrometer handle (Fig. 6). Similar designs are available from Rhein Medical, Storz, ASICO, and other manufacturers.
Another less common method of creating peripheral relaxing incisions is to use a device such as the Terry/Schanzlin Astigmatome (Oasis Medical), which circumvents the need to make a “free-hand” incision. This trephine-like device has been designed to create consistent and symmetric peripheral arcuate corneal relaxing incisions. It uses a vacuum speculum that mates with various reusable templates that are selected based upon the amount of astigmatic correction that is desired. The incision is created by simply rotating a disposable steel blade unit that fits inside of the template (Fig. 7).
As discussed, LRIs are proving to be a safer and more forgiving approach to managing astigmatism as compared to more central corneal incisions. Nonetheless, as with any surgical technique, potential complications exist, and several are listed in Table 3. Of these, the most likely to be encountered is the placement of incisions upon the wrong axis. When this occurs, it typically takes the form of a 90-degree error with positioning upon the opposite, flat meridian. This, of course, results in an increase and likely doubling of the patient's preexisting cylinder. Compulsive attention is required in this regard. The surgeon ought to consider employing safety checks to prevent this frustrating complication from occurring, such as having a written plan that is brought into the operating room, kept visible and properly oriented. Incisions are always placed upon the plus (+) cylinder axis and opposite to the minus (−) cylinder axis.
Although very rare when utilizing a blade depth setting of 600 microns, corneal perforation is possible. This may be due to improper setting of the blade depth or as a result of a defect in the micrometer mechanism. This latter problem may arise after repeated autoclaving and many sterilization runs. Periodic inspection and calibration is therefore warranted, even with preset single depth knives. When encountered, unlike radial microperforations, these circumferential perforations will rarely self-seal and will likely require placement of temporary sutures.
As mentioned, LRIs lend themselves well to in-office “touch-ups.” Although some surgeons will place or extend incisions at the slit-lamp, it is our preference to use a small operating microscope and to perform the procedure within a dedicated treatment room. It has been our experience that this provides far better surgical control as well as patient comfort. In the case of residual astigmatism without prior incisional correction, one uses the same technique and nomogram as described above.
In the case of an undercorrection following previous LRIs, one should inspect the length and positioning of the incisions. As indicated, placement of the incisions too far out into the true surgical limbus and beyond clear cornea will often lead to undercorrection. If the arc length and location appear to be adequate, one ought to suspect that the patient has an unusually thick cornea. This occurs most frequently in hyperopic eyes. In this situation, pachymetry should be performed and the incisions may be redeepened or extended. When faced with an overcorrection, one should resist the temptation to place additional incisions in the opposite meridian. This can lead to an unstable cornea with unpredictable refractive results, or worse, induce irregular astigmatism. Rather, one should consider nonincisional modalities such as PRK or LASIK. We also have had good results in this setting using conductive keratoplasty off-label, particularly if the overcorrection involves hyperopic astigmatism.29
To correct unusually high levels of astigmatism, LRIs may be used in conjunction with a toric IOL or excimer laser surgery (bioptics). In several rare cases, we have combined all three modalities and safely corrected up to 9 D of preexisting astigmatism!
|Case 1 is a 68-year-old male who presented for right cataract surgery. His
refraction was −1.00 +2.25 × 80 and was recorded
as reliable, consistent with his modest cataract density. Keratometry
readings were 44.75 × 75 and 43.00 × 165. Corneal topography
confirmed slightly more than 2.00 D of regular and slightly oblique
cylinder. Consulting the nomogram, a plan was devised for a pair of LRIs
to be centered over the 75-degree axis, with each incision delineating 45 degrees
of arc. A single plane phaco incision was used and maintained
at a size of less than 3.2 mm (Figs. 8–11).|
Case 2 is a 79-year-old woman who presented with a very dense left cataract. Her refraction was recorded at −2.25 +2.75 × 125 with a difficult end point. Her manual keratometry and topography measurements were consistent and revealed slightly less than 1.75 D at 120 degrees. Because of the questionable refraction, greater value was placed on the corneal measurements. Based upon the cataract nomogram, the plan was for paired LRIs of 40 degrees to be placed over the steep 120-degree axis (Figs. 12–19).
Case 3 is a 48-year-old bilateral hyperope who presented for a refractive surgical consultation. The refraction in his left eye was found to be +3.25 +1.75 × 85. Keratometry was somewhat flat but confirmed WTR cylinder as did corneal topography. Based upon the patient's age, refraction, and somewhat shallow anterior chambers, the decision was made to proceed with a refractive lens exchange. The NAPA nomogram called for LRIs of 55 to 60 degrees with intraoperative pachymetry. Intraoperative keratoscopy confirmed the steep 85-degree meridian (Figs. 20–29).
17. Osher RH: Combining phacoemulsification with corneal relaxing incisions for reduction of preexisting astigmatism. Paper presented at Annual meeting of the American Intraocular Implant Society, Los Angeles, CA, 1984
18. Maloney WF: Refractive cataract replacement: a comprehensive approach to maximize refractive benefits of cataract extraction. Paper presented at Annual Meeting of the American Society of Cataract and Refractive Surgery, Los Angeles, CA, 1986
19. Osher RH: Transverse astigmatic keratotomy combined with cataract surgery. In Contemporary Refractive Surgery-Ophthalmology Clinics of North America. Thompson K, Waring G, Eds. Philadelphia, PA: WB Saunders, 1992:717
22. Nichamin LD: Changing approach to astigmatism management during phacoemulsification: Peripheral arcuate astigmatic relaxing incisions. Paper presented at Annual Meeting of the American Society of Cataract and Refractive Surgery, Boston, MA, 2000
30. Schimmelpfenning BH, Waring GO: Development of radial keratotomy in the nineteenth century. In Refractive keratotomy for myopia and astigmatism. Waring GO, ed. St. Louis, MO: Mosby-Year Book, 1992:174