Astigmatic Correction

Ophthalmologists discuss techniques and devices used to correct
astigmatism as well as postoperative quality of vision.
In Accurate Astigmatic Keratotomy
Robert M. Kershner, MD, MS, FACS

In 1994, I published my results with a technique involving clear corneal cataract surgery and the simultaneous correction of myopia, hyperopia, and astigmatism. I called this approach keratolenticuloplasty to reflect the surgical modification of the cornea combined with the correction of lenticular error by replacing the natural lens with an IOL.1,2 My results demonstrated that surgeons could improve UCVA with the cataract procedure. The development of new lens designs, wavefront analysis, and even better surgical instruments and techniques has brought the science of vision correction and neutralization of refractive error during the cataract procedure to a new level.3-6 This article focuses on how to combine modern refractive correction with cataract surgery. After all, what good is an accommodating lens to a patient who must wear spectacles for astigmatism?


I have been amazed, and a bit disappointed, that cataract surgeons as a group were slow to begin correcting their patients’ refractive errors. After the techniques were proven and the necessary instruments were available from enterprising, innovative manufacturers, patients still had to wear spectacles to correct their astigmatism and presbyopia after cataract surgery.

Ophthalmologists’ attitudes toward eliminating or reducing patients’ need for spectacle and contact lens correction began to change slightly when the late Svyatoslav Fyodorov, MD, introduced radial keratotomy. Even with today’s laser technology, incisional keratotomy still provides the simplest and most reproducible approach to astigmatic correction during cataract surgery.

Contemporary patients now demand and should expect unaided, clear postoperative vision. The push for emmetropia with phakic, aphakic bifocal, and accommodative IOLs has finally created an impetus for all surgeons to eliminate ametropia, especially astigmatism, after cataract surgery. Eighty-nine percent of patients can achieve adequate visual acuity without spectacles for most tasks using today’s surgical techniques.4


In all patients, surgeons should fully correct the sphere for distance and eliminate more than 0.75D of astigmatism with a single-incision keratotomy that can also act as the IOL’s insertion site. The goal should be full correction or a slight undercorrection of the cylinder, never an overcorrection or shift in the cylinder’s axis. To avoid the possibility of error, I create a preoperative surgical plan.

Limbal Relaxing Incisions

The term limbal relaxing incision (LRI) is a misnomer. The flattening incisions made in the cornea are neither relaxing nor limbal. Where one makes the incision is just as important as how one makes it. Prior to surgery, I note in the chart the position of the patient’s steepest corneal meridian (I recommend consulting a worksheet). The surgeon should carefully evaluate the patient’s preoperative astigmatism and take a topographic map into the OR. Corneal incisions can consistently and predictably alter the corneal curvature. First described by Spencer Thornton, MD, of Nashville, Tennessee, one of the early American pioneers in refractive keratotomy, all transverse or arcuate corneal incisions will flatten the cornea in the meridian in which they are placed by acting as if tissue were added to the keratotomy site.

The Best Flattening Incision

I have long championed a single, small, arcuate incision (AK) on near clear cornea at an optical zone of approximately 10mm. Small incisions avoid trouble, because their flattening effect occurs only in the meridian where they are placed while the cornea becomes steeper 90º away through a process known as coupling. If a single incision is insufficient, I simply pair it with another cut on the opposite meridian. Rather than flatten only the steep meridian, large incisions have a more global effect, and they increase the potential for irregular astigmatism and a change in the overall corneal power.

How do arcuate astigmatic incisions differ from LRIs? Because they are placed far peripherally in the cornea at the scleral limbus and farthest from the optical center of the eye, LRIs have a less flattening effect for a given length. As a result, they must be large to substantially affect the corneal curvature. When LRIs exceed 120º of arc, especially if they are placed nasally or temporally, they damage the corneal nerves at that location and create problems with dry eye and healing. Smaller, arcuate incisions (usually 3mm) that closely follow the natural corneal curvature produce a greater effect with less cutting and little risk compared with LRIs. I recommend following a published nomogram, using the appropriate tools, and making the incision deep enough (at least 85% to 95% corneal thickness) to have a permanent effect. The results will be highly predictable (Figure 1).


I have been pleased to see the large number of AK instruments now available from numerous manufacturers. When performing AK, I administer topical anesthesia so that the patient can fixate on the OR microscope’s light. For safety’s sake, I use a Thornton-style ring to fixate the globe. I mark the location of the AK on the steepest meridian using one of the astigmatic markers (Rhein Medical, Inc., Tampa, FL) I designed. For surgeons who prefer the sharpness of the diamond blade and have the budget and staff to purchase and maintain one, I would recommend the Seibel LRI Diamond Knife (Rhein Medical, Inc.), the Kershner AK blade (Diamatrix Ltd., The Woodlands, TX), or the Thornton Arc T Blade (Mastel Precision, Inc., Rapid City, SD). Diamond blades are best for a smooth, reproducible, full-depth incision of greater than 85% to 90%. For diamond-like quality in a disposable product, I have been impressed with the BD Atomic Edge Blade (BD Ophthalmic Systems, Waltham, MA), which is made from silicon semiconductor material. The products I have mentioned are only a sampling of the many available, and more are developed every day.

I suggest taking a pachymetry reading at the site of the AK and then setting the blade to 100% of this measurement. The most common cause of inadequate correction is an insufficient depth to ensure flattening. If a pachymeter is not available, the ophthalmologist may use a preset blade or manually set the blade to between 600 and 650µm.


Investing time and money in learning AK, a procedure that is usually not reimbursed, repays the surgeon many times over with better outcomes, a benefit for patients as well. As new IOL technologies that restore accommodation and enhance visual function gain increasing acceptance among ophthalmologists, the need to deliver clear UCVA becomes more important than ever. This goal is within the reach of every refractive cataract surgeon.

Published nomograms are available at and at

Robert M. Kershner, MD, MS, FACS, is President and CEO of Eye Laser Consulting in Boston. He states that he holds no financial interest in any product or company mentioned herein. Dr. Kershner may be reached at

1. Kershner RM. Refractive Keratotomy for Cataract Surgery and the Correction of Astigmatism. Thorofare, NJ: Slack, Inc.; 1994.

2. Kershner RM. Keratolenticuloplasty. In: Gills JP, Sanders DR, eds. Surgical Treatment of Astigmatism. Thorofare, NJ: Slack, Inc.; 1994: 143-155.

3. Kershner RM. Keratolenticuloplasty: arcuate keratotomy for cataract surgery and astigmatism. J Cataract Refract Surg. 1995;21:274-277.

4. Kershner RM. Clear corneal cataract surgery and the correction of myopia, hyperopia and astigmatism. Ophthalmology. 1997;104:381-389.

5. Kershner RM. Optimizing the refractive outcome of clear cornea cataract surgery. In: Agarwal S, Agarwal A, Agarwal A, eds. Phaco, Phakonit and Laser Phaco—a Quest for the Best. Dorado, Republic of Panama: Highlights of Ophthalmology; 2002: 85-104.

6. Kershner RM. Refractive keratotomy and the toric IOL for the correction of astigmatism in clear cornea cataract surgery. In: Gills J, ed. A Complete Surgical Guide for Correcting Astigmatism. Thorofare, NJ: Slack, Inc.; 2002: 49-64.

The Staar Toric IOL
David F. Chang, MD

To address preexisting astigmatism at the time of cataract surgery, I prefer limbal relaxing incisions (LRIs) because they are quick and virtually free of complications.

There are two categories of patients, however, in whom LRIs are less effective and for whom I have been using a toric IOL instead. One is younger cataract patients (< 60 years) with more than 2.00D of astigmatism, because a younger age significantly reduces the attainable effect from incisional keratotomy. The second group includes patients with the highest amounts of astigmatism (eg, > +3.00D). LRIs are much less predictable in these eyes, and, for against-the-rule cylinder, I want to avoid combining my temporal clear corneal incision within a large, temporal LRI. This article highlights data on and my own experience with the Staar Toric IOL (Staar Surgical Company, Monrovia, CA).


The FDA approved the Staar silicone plate-haptic toric IOL in November 1998, and it remains the only available toric IOL in the US today. The IOL is available in two astigmatic powers; the +2.00D toric lens corrects 1.50D of keratometric astigmatism, and the +3.50D toric power corrects approximately 2.25D at the spectacle plane. I exclusively use the +3.50D cylindrical correction to address those cases where LRIs are less effective. The Staar Toric IOL is also available in two different lengths: the 10.8-mm–long TF model, which was the original FDA-studied design, and the 11.2-mm TL model, released in 1999, for spherical powers of ≤ 23.50D.


A primary risk with any toric IOL is postoperative off-axis rotation. Any misalignment of the toric axis decreases the amount of astigmatism reduced. With 10º of axis deviation, approximately one-third of the effect is lost. With 20º of axis deviation, approximately two-thirds of the effect is lost. Misalignment of the lens greater than 30º produces a net worsening of astigmatism. With the Staar Toric IOL, late rotation has not been a problem, suggesting that, once the capsular bag has fully contracted, the torsional fixation is permanent. Early postoperative rotation of the TF model has been a significant issue, however.

According to FDA study data, 24% of the toric IOLs ended up more than 10º off axis: 12% were >20º off; 8% were >30º off; and 5% were >45º off axis. The FDA study evaluated only the shorter 10.8-mm TF lens. Subsequently, a number of published clinical studies1-4 of the Staar Toric IOL confirmed a significant incidence of more than 10º to 15º of rotation postoperatively with the shorter TF model (Table 1).

Staar Surgical Company’s release of the longer TL model was an effort to decrease early rotation. When the longer toric IOLs became available in 1999, I undertook my own study of the rotational stability of this model.5 All of the earlier published reports had studied the TF model, which was the only available model at that time. At the 2003 ASCRS meeting, I updated my results from an expanded study of 90 consecutive Staar Toric IOL implantations.6 All 90 implants were with the higher +3.50D toric power. Eighty were the longer TL IOLs, and 10 were TF IOLs in powers > 24.00D (where the TL model is not available).

The IOLs exhibited excellent rotational stability. In the 80 consecutive TL toric implants, 73% were within 5º, 89% were within 10º, and 96% were within 15º of the target axis. These results represented a significant improvement compared with both the FDA data and the data from previously published series1-4 in which the shorter TF model was used. My repositioning rate was 2.5% (two of 80 lenses) with the longer TL model and 3.3% overall (one of 10 TF models).


I continue to employ the same surgical guidelines for implanting the Staar Toric IOL that were used during my study.5,6 I always use the longer TL model if it is available (power < 23.50D). I inflate the capsular bag with a cohesive viscoelastic, because dispersive viscoelastics render the silicone IOL surface more slippery. To maximize the contact between the IOL and the posterior capsule, I remove any viscoelastic trapped behind the optic with the I/A handpiece. I try not to overly inflate the eye, because leaving the globe somewhat softer probably allows the flaccid capsular bag to collapse around the IOL more quickly.

I make an astigmatically neutral, temporal, clear corneal incision and employ topical anesthesia without postoperative patching or shields. While the patient is sitting upright on the operating table just prior to surgery, I dot the limbus at the 6-o’clock position with a skin-marking pen. Beneath the microscope, a Mendez-style degree gauge facilitates the orientation of two limbal ink marks that identify the desired axis. After implanting the IOL, I can align its axis marks with these ink marks (Figure 2) and double check them against the preoperative notes or chart.

Any repositioning procedure should be performed within the first postoperative week. Once the capsular bag fully contracts, rotating the IOL requires greater force, which might increase the chance of tearing the capsular bag or zonules. Plate haptic IOLs should never be implanted in the sulcus or in the presence of a torn capsulorhexis or posterior capsule.


Toric IOLs are an excellent complement to corneal astigmatic incisions, and they are particularly useful for patients with very high amounts of keratometric cylinder. Early clinical studies indicate excellent rotational stability with the toric Acrysof IOL (Alcon Laboratories, Inc., Fort Worth, TX) as could be expected due to its tacky acrylic material. Because of its advantages over a plate haptic silicone design, I believe that the new Acrysof will become the toric IOL of choice once approved by the FDA. For the past 5 years, however, I believe that the Staar Toric IOL has been grossly underutilized in part because of ophthalmologists’ misconceptions about its true rate of rotation. Proper surgical technique and the use of the longer lens model have provided excellent rotational stability and efficacy. Until a better toric IOL comes along, the Staar Toric IOL is an excellent and viable option for cases in which LRIs are least effective.

David F. Chang, MD, is Clinical Professor of Ophthalmology at the University of California, San Francisco and is in private practice iin Los Altos, California. He states that he holds no financial interest in any product or company mentioned herein. Dr. Chang may be reached at (650) 948-9123;

1. Sun XY, Vicary D, Montgomery P, Griffiths M. Toric intraocular lenses for correcting astigmatism in 130 eyes. Ophthalmology. 2000;107:1776-1781; discussion by Kershner RM: 1781-1782.

2. Ruhswurm I, Scholz U, Zehetmayer M, et al. Astigmatism correction with foldable toric intraocular lens in cataract patients. J Cataract Refract Surg. 2000;26:1022-1027.

3. Leyland M, Zinicola P, Bloom P, Lee N. Prospective evaluation of a plate haptic toric intraocular lens. Eye. 2001;15:202-205.

4. Till JS, Yoder PR, Wilcox TK, Spielman JL. Toric intraocular lens implantation: 100 consecutive cases. J Cataract Refract Surg. 2002;28:295-301.

5. Chang DF. Early rotational stability of the longer Staar Toric IOL–50 consecutive (TL) IOLs. J Cataract Refract Surg. 2003;29:935-940.

6. Chang DF. Early rotational stability of the longer Staar Toric IOL: 70 consecutive cases. Paper presented at: The ASCRS/ASOA Symposium on Cataract, IOL and Refractive Surgery; April 14, 2003; San Francisco, CA.
New-Generation Toric IOL
Warren E. Hill, MD

At present, cataract surgeons have only two options for correcting significant corneal astigmatism: limbal relaxing incisions (LRIs), or a plate haptic toric intraocular lens.

Of course, LRIs remain an old and trusted techniques, which work well to lessen low magnitudes of corneal astigmatism. They are easy to perform and generally do not require a great deal of precision. However, LRIs cannot be used in all situations. Although the first-generation toric IOL gave us another useful option, it also presented some challenges—mostly with rotational stability and a limited power range. The next generation of toric IOLs offers refinements that elegantly address these concerns.

The Acrysof Toric IOL (Alcon Laboratories, Inc., Fort Worth, TX), which is currently undergoing FDA clinical trials, is based on the familiar SA60 single-piece acrylic platform (Alcon Laboratories, Inc.). Aside from an expanded power range and excellent rotational stability, this newest addition to our surgical armamentarium offers another refinement to its use: the surgeon will now be able to take into account the astigmatism induced by the cataract surgical wound (Figure 3). This approach results in an enhanced prediction of the true axis of postoperative corneal astigmatism and a more accurate prediction of the cylindrical power required.


Any measurement that has both magnitude and direction can be considered a vector. For example: +2.00D of corneal astigmatism at an axis of 135º would be a vector quantity. This measurement becomes important when you stop to consider the fact that both the axis and the magnitude of corneal astigmatism prior to cataract surgery are often different than after cataract surgery. This is especially true if the incision is large or is placed superiorly.

Although many surgeons no longer think about modern clear corneal cataract incisions’ inducing astigmatism, for most corneal incisions, the vector of the corneal wound invariably changes the vector of the corneal astigmatism, depending on the location, size, and architecture. If the toric IOL power selection and axis of placement are solely based on preoperative keratometry, an astigmatic angular error may occur, even in the absence of IOL rotation. Those who have implanted first-generation toric IOLs, where the effect of the wound is completely ignored, are all too familiar with this finding, as both the axis and the amount of refractive astigmatism often do not match what would be expected.

Because corneal astigmatism can be viewed as a vector, and the astigmatism induced by the corneal wound is also a vector, these two vectors can be added together. Just as a pilot may be required to change power and course direction to compensate for the wind, the surgeon may also be required to calculate a new toric IOL power and a new toric IOL axis to compensate for astigmatic changes induced by the cataract wound.

For example, a cataract surgery patient has a low degree of astigmatism, say +1.25D, to be corrected by a toric IOL, with the steep axis at 135º. The surgeon makes either a single- or double-plane temporal clear corneal incision at 180º, which induces approximately 0.50D of steepening at 90º. Although this change may not appear significant, ignoring the resultant axis shift will produce an angular error of approximately 0.50D at the corneal plane and approximately 0.75D at the plane of the capsular bag. In other words, the toric power of the IOL would be off by 0.75D.

As another example, a cataract patient has +2.00D of corneal astigmatism to be corrected by a toric IOL, with the steep axis at 135º. The surgeon makes a scleral tunnel incision at 90º, which, for this particular surgeon, typically induces 0.75D of steepening at 180º. Even though the amount of corneal astigmatism has not significantly changed, ignoring the axis shift induced by this incision would produce a residual angular error of approximately 0.75D at the corneal plane and approximately 1.00D at the plane of the capsular bag. In other words, the toric power of the IOL would be off by +1.00D.

Not only will the Acrysof Toric IOL be on a rotationally stable platform, but also for the first time, companion software will be made available to surgeons to more precisely calculate the new resultant axis and power of astigmatism, based on their individual incision data.

My partner, Neal Nirenberg, MD, and I were investigators for the recently completed phase III FDA study of 500 eyes implanted with this toric technology. The data from this study revealed exceptional rotational stability with 83% of implantations remaining within 5º of the target axis. The 100-day data from this same study showed that approximately 80% of patients had less than 0.75D of residual postoperative astigmatism, and close to 70% had less than 0.50D of residual postoperative astigmatism. These results confirmed our initial impression that the addition of vector analysis as a further refinement to toric IOL power calculations is a valuable exercise.


In summary, this newest generation of toric IOLs exhibits excellent rotational stability and can be used with companion vector analysis software to further optimize refractive accuracy. This approach requires careful preoperative keratometry and an awareness of the amount of astigmatism induced by each surgeon’s typical incision. The new Acrysof Toric IOL will be the first lens to employ such a strategy.

Warren E. Hill, MD, FACS, is Medical Director of East Valley Ophthalmology in Mesa, Arizona. He has worked as a consultant in the area of IOL mathematics for Alcon Laboratories, Inc. Dr. Hill may be reached at (480) 981-6130;

Bioptics and Refractive IOLs
John Doane, MD, FACS

What is bioptics, why do we employ this technique, and how do we use it in a planned and an unplanned fashion? Many surgeons currently use bioptics to achieve emmetropia, whether with a combination of a refractive IOL and a conventional IOL in sulcus piggyback fashion or, alternatively, a refractive IOL (phakic or pseudophakic) combined with laser vision correction to achieve the refractive target.


During a planned bioptics procedure, the surgeon may first implant a phakic refractive lens or a pseudophakic lens followed by corneal laser vision correction, or he may place a pseudophakic lens and follow this with a conventional lens in sulcus piggyback fashion. Unplanned bioptics is considered to be the enhancement of any refractive or pseudophakic IOL with a laser procedure or sulcus piggyback IOL. For example, the treatment of a patient who has 5.00D of corneal astigmatism requires the combination of two different procedures (eg, combination of a toric IOL and limbal relaxing incision[s] or a combination of a toric or conventional IOL and corneal laser vision correction).

An early example of planned bioptics is what occurred in the early 1990s with automated lamellar keratoplasy (ALK) combined with radial keratotomy. Typically, ALK for myopia would leave residual myopia of 1.00 to 3.00D. After 3 months or so, the surgeon could treat the patient with radial keratotomy to achieve near emmetropia. During the past 5 years, international ophthalmologists who have been implanting phakic IOLs have combined the lenses with corneal laser vision correction to treat any residual spherical refractive error or preexisting cylindrical error. In the past year or so, I have used the Crystalens (Eyeonics, Inc., Aliso Viejo) in bioptics fashion with either conventional IOLs in the ciliary sulcus or laser vision correction to treat residual spherical error or preexisting corneal cylindrical error.


In a planned bioptics technique, a surgeon can implant the Crystalens or multifocal IOLs bilaterally and wait up to 3 months to allow the corneoscleral wound to heal before performing a lamellar procedure, such as LASIK, that raises the IOP. The surgical interval in these cases is approximately 12 to 14 weeks. I term this interval refractive purgatory because the surgeon must fit the patient with spectacles or contact lenses to make his vision functional for work, driving, and outside activities. Alternatively, a surgeon can create lamellar flaps, implant accommodative or multifocal lenses in both eyes, wait 4 weeks for the lamellar flap and the corneoscleral wounds to settle to refractive stability, and then lift the corneal flap and ablate the residual refractive error with an excimer laser. Lastly, if a surgeon prefers surface ablative techniques he can implant the IOL and allow for sufficient postoperative refractive stability (3 to 4 weeks) and then proceed with surface excimer laser treatment.


Cataract/refractive surgeons are commonly faced with a refractive or cataract case that requires an accommodative IOL power that is outside the available range, or surgeons need to adjust for a refractive surprise. In some of these cases, I will place a piggyback IOL in the ciliary sulcus. I have, been using the Staar AQ5010V (Staar Surgical Company, Monrovia, CA), which comes in the range of -4.00 to +4.00D, and the Clariflex IOL (Advanced Medical Optics, Inc., Santa Ana, CA), which comes in low dioptric powers in 0.50D increments. I use the AQ5010V or the Clariflex low plus lenses for patients who are left with lower amounts of hyperopic refractive error (planned or unplanned). I allow for a 1-month lapse in between implanting the sulcus piggyback lens to ensure a stable refraction. Other surgeons may opt for excimer hyperopic ablation or conductive keratoplasty to achieve emmetropia. Personally, I prefer the refractive stability of an IOL.

A typical scenario in my experience is that a patient presents with a large amount of corneal cylinder or requires an IOL power that is not available. For example, the Crystalens is available in powers ranging from +10.00D to +33.00D, but some patients may need a +2.00, +35.00, or even a +38.00D lens. For highly myopic and/or astigmatic patients, I create bilateral corneal flaps, bilaterally implant the Crystalens, wait 4 weeks, lift the flap, and then ablate any residual refractive error. Alternatively, for extremely myopic patients, I place negative power IOLs in the sulcus to achieve near emmetropia and take advantage of accommodative IOL technology in the capsular bag. To date, I have not found that an IOL in the sulcus negatively affects the accommodative ability of the Crystalens. Conversely, in cases of very short eyes with high hyperopia, I may implant a +33.00D Crystalens followed by an appropriately powered plus lens in the ciliary sulcus to achieve near emmetropia.


In cases as described earlier, the only option for individuals with refractive errors that are far from the mean refractive error of the general population is to undergo more than one surgical procedure. From this viewpoint, bioptics is the only modality to allow this patient population the opportunity to take advantage of the latest ophthalmic device technologies. Ideally, a single procedure would suffice, but at present the surgeon and patient must accept the reality of the status quo and plan appropriately to achieve the desired goals.
Bioptics will likely continue to be a process that certain patients can take advantage of in a planned or unplanned fashion. The important point to remember is that refractively minded patients expect nearly perfect vision, which equates to near emmetropia; therefore, the ophthalmic surgeon interested in refractive IOL surgery will have to be well versed in the role and benefits of bioptic techniques.

John F. Doane, MD, FACS, is in private practice with Discover Vision Centers in Kansas City, Missouri, and is Clinical Assistant Professor for the Department of Ophthalmology, Kansas University Medical Center. Dr. Doane is a co-course director of Eyeonics Crystalens certification courses and has been compensated by Eyeonics for research overhead during the FDA clinical investigation of the Crystalens. Dr. Doane may be reached at (816) 478-1230;
For a downloadable pdf of this article, including Tables and Figures, click here.