Pearls of Supracapsular Surgery Using the Tilt and Tumble Technique1

The technique of tilt and tumble is a modified form of supracapsular phacoemulsification. It uses a bimanual technique to tilt one pole of the nucleus above the anterior capsule. Phacoemulsification is then performed while supporting the lens in the iris plane with a nucleus rotator.

In the following paragraphs we will describe and illustrate this technique in enough detail to allow an ophthalmologist to perform it on their own.

The indications for the tilt and tumble phacoemulsification technique are quite broad. It can be utilized in either a large or small pupil situation. Some surgeons favor the technique with small pupils where the nucleus can be tilted up such that the equator is resting in the center of the pupil and is then carefully emulsified. It does require a larger continuous tear anterior capsulectomy of at least 5.0 mm. If a small anterior capsulectomy is created, the hydrodissection step of tilting the nucleus can be dangerous, and it is possible to rupture the posterior capsule during the hydrodissection step. If, inadvertently, a small anterior capsulectomy is created, it is probably safest to convert to an endocapsular phacoemulsification technique or enlarge the capsulorhexis. If it is not possible to tilt the nucleus with either hydrodissection or a manual technique, the surgeon should also convert to an endocapsular approach. Occasionally the entire nucleus will subluxate into the anterior chamber. In this setting if the cornea is healthy, the anterior chamber 2deep, and the nucleus soft, then the phacoemulsification can be completed in the anterior chamber supporting the nucleus away from the corneal endothelium. The nucleus can also be pushed back inferiorly over the capsular bag to allow the iris plane tilt and tumble technique to be completed.

In patients with severely compromised endothelium, such as Fuchs’ dystrophy or previous keratoplasty patients with a low endothelial cell count, endocapsular phacoemulsification is preferred to reduce endothelial cell loss. In a normal eye, corneal clarity on the first day postoperatively is excellent. Nevertheless, the tilting and tumbling maneuvers do increase the chance of endothelial cell contact of lens material compared to an endocapsular phacoemulsification. Therefore, the endocapsular technique should not be employed in eyes with borderline corneas or shallow anterior chambers.

The patient enters the anesthesia induction or preoperative area and Tetracaine drops are placed in both eyes. The placement of these drops increases the patient comfort during the placement of the multiple dilating and preoperative medications, decreases blepharospasm and also increases the corneal penetration of the drops to follow.

The eye is dilated with 2.5% neosynephrine and 1% cyclopentolate every 5 minutes for three doses. Additionally, preoperative topical antibiotic and antiinflammatory drops are administered at the same time as the dilating drops. We favor the combination of a preoperative topical antibiotic, topical steroid and topical non-steroidal. The rationale for this is to pre-load the eye with antibiotic and non-steroidal prior to surgery. The pharmacology of these drugs and the pathophysiology of postoperative infection and inflammation support this approach. An eye that is pre-loaded with anti-inflammatories prior to the surgical insult is likely to have a much reduced postoperative inflammatory response. Both topical steroids and non-steroidals have been found to be synergistic in the reduction of postoperative inflammation. In addition, the use of perioperative antibiotics is supported in the literature as reducing the small chance of postoperative endophthalmitis. Since the patient will be sent home on the same drops utilized preoperatively, there is no additional cost.

Our usual anesthesia is topical tetracaine reinforced with intraoperative intracameral 1% non-preserved (methylparaben free) xylocaine. For patients with blepharospasm a “miniblock” O’Brien facial nerve anesthesia, utilizing 2% xylocaine with 150 units of hyaluronidase per 5 cc of xylocaine, can be quite helpful in reducing squeezing. This block lasts thirty to forty-five minutes and makes surgery easier for the patient and the surgeon. Patients are sedated prior to the block to eliminate any memory of discomfort. One way to determine when this facial 3nerve block might be useful is to ask the technicians to make a note in the chart when they have difficulty performing applanation pressures or A-scan because of blepharospasm. In these patients a mini facial nerve block can be quite helpful. In younger anxious patients and in those with difficulty cooperating, we perform a peribulbar block. Naturally, general anesthesia is used for very uncooperative patients and children. While this is controversial, in some patients where general anesthesia is chosen and a significant bilateral cataract is present, we will perform consecutive bilateral surgery completely re-prepping and starting with fresh instruments for the second eye. Again, this is a clinical decision weighing the risk to benefit ratio of operating both eyes on the same day versus the risk of two general anesthetics.

Upon entering the surgical suite the patient table is centered on pre-placed marks so that it is appropriately placed for microscope, surgeon, scrub nurse and anesthetist access. We favor a wrist rest, and the patient's head is adjusted such that a ruler placed on the forehead and cheek will be parallel to the floor. The patient's head is stabilized with tape to the head board to reduce unexpected movements, particularly if the patient falls asleep during the procedure and suddenly awakens. A second drop of tetracaine is placed in each eye. If the tetracaine is placed in each eye, blepharospasm is reduced. A periocular prep with 5% povidone-iodine solution is completed. We do not irrigate the ocular surface and fornices with povidone-iodine. Under topical anesthesia we have found that the patients note a significant burning. If a few drops leak into the eye this is certainly acceptable.

An aperture drape is helpful for topical anesthesia to increase comfort. We have noted that when the drape is tucked under the lids this often irritates the patient's eye and also reduces the malleability of the lids, decreasing exposure. Since it is important to isolate the meibomian glands and lashes a Tegaderm adhesive cut in half for the upper and lower lids may be used.

Balanced salt solution is used in all cases. For the short duration of a phacoemulsification case, BSS plus does not provide any clinically meaningful benefit. We place 0.5cc of the intracardiac non-preserved (sodium bisulfate free) epinephrine in the bottle for assistance in dilation and perhaps hemostasis. We also add 1ml (1,000 units) of heparin sulfate to reduce the possibility of postoperative fibrin. This is also a good anti- inflammatory and coating agent. At this dose there is no risk of enhancing bleeding or reducing hemostasis.

The lids are separated with a Lindstrom/Chu aspirating speculum (Rhein Medical). A final drop of tetracaine is placed in the operative eye or the surface is irrigated with the non-preserved xylocaine. We do not like to utilize more than three drops of tetracaine or other topical anesthetic as excess softening of the epithelium can occur, resulting in punctate epithelial keratitis, corneal erosion and delayed postoperative rehabilitation.4

The patient is asked to look down. The globe is supported with a dry Merocel sponge, and a counter puncture is performed superiorly at 12 o'clock with a diamond stab knife. (Osher/ Storz) The incision is about 1 mm in length (Figure 1). Approximately 0.25 ml of 1% non-preserved methylparaben free xylocaine is injected into the eye (Figure 2). We advise the patient that they will feel a “tingling” or “burning” for a second, and then “the eye will go numb”. This provides a psychological support for the patient that they will now have a totally anesthetized eye and should not anticipate any discomfort. We tell them that while they will feel some touch and fluid on the eye, they will not feel anything sharp, and if they do, we can supplement the anesthesia. This injection also firms up the eye for the clear corneal incision. We do not find it necessary to inject viscoelastic prior to constructing the corneal wound.

We perform a temporal or nasal anterior limbal or posterior clear corneal incision. Care is taken not to incise the conjunctiva as this can result in ballooning during phacoemulsification and irrigation aspiration. Some surgeons define this as being a posterior clear corneal incision and others as an anterior limbal incision. The anatomical landmark is the perilimbal capillary plexus and the insertion of the conjunctiva. Since the incision is into a vascular area, long-term wound healing can be expected to be stronger than it is with a true clear corneal incision. True clear corneal incisions, such as performed in radial keratotomy, clearly do not have the wound healing capabilities that a limbal incision demonstrates where there are functioning blood vessels present.

The anterior chamber is then entered parallel to the iris at a depth of approximately 300 microns. This creates a hinge type of incision (Figure 3).

In right eyes the incision is temporal, and in left eyes, nasal. This allows the surgeon to sit in the same position for right and left eyes. The nasal cornea is thicker, has a higher endothelial cell count and allows very good access for phacoemulsification. The nasal limbus is approximately 0.3 mm closer to the center of the cornea than the temporal limbus, and this can, in some cases where there is excess edema, reduce first day postoperative vision more than one might anticipate with a temporal incision. There also can, in some patients, be pooling of irrigating fluid. For this reason, an aspirating speculum is useful. It is also helpful to tip the head slightly to the left side. Nonetheless, in left eyes a nasal clear corneal approach is an excellent option, particularly for surgeons who find the left temporal position uncomfortable.6

In some patients it may be safest to create a corneal scleral incision. Examples of these include patients who have had a previous radial keratotomy or demonstrate findings of peripheral corneal ulcerative keratitis, in some patients with very low endothelial cell counts, and any case where there is any significant peripheral pathology or thinning. The anterior limbal or posterior corneal incision described above can be made temporally, nasally, in the oblique meridian or even superiorly without induction of significant corneal edema or endothelial cell loss.

The incision, if 3 mm in length, tends to cause an induction of 0.25 ± 0.25 diopters of astigmatism. If it is placed on the steeper meridian, it can therefore be expected to reduce the astigmatism somewhere between 0 and 0.50 diopters. An incision in the 3 mm range will almost always be self sealing.

With modern injector systems most foldable intraocular lenses can be implanted through a 3 mm anterior limbal incision.

In select patients an intraoperative astigmatic keratotomy can be performed at the 7 to 8 mm optical zone. This can be done at the beginning of the operation. The patient's astigmatism axis is marked carefully using an intraoperative surgical keratometer which allows one to delineate the steeper and flatter meridian and not be concerned about globe rotation. One 2 mm incision at a 7 to 8 mm optical zone will correct 1 diopter of astigmatism and two 2 mm incisions will correct 2 diopters of astigmatism in a cataract age patient. One 3 mm incision will correct 2 diopters, and two 3 mm incision 4 diopters. One can combine a 3 and 2 mm correcting 3 diopters. Larger amounts of astigmatism can also be corrected utilizing the Arc-T nomogram. Depending on the age of the patient one can correct up to 8 diopters of astigmatism with two 90° arcs. Many surgeons have moved to a more peripheral corneal limbal arcuate incision, but we favor the 7 – 8 mm optical zone because of years of experience with this approach. There certainly is a variation in response, but there have not been any significant induced complications with this approach. The outcome goal is 1 diopter or less of astigmatism in the preoperative axis. It is preferable to under-correct rather than over-correct. The key in astigmatism surgery is “axis, axis, axis”. If one is not careful in preoperative planning and the incision are placed more than 15° off axis, one is better avoiding this approach. The anterior chamber is constituted with a viscoelastic. Our studies have not found any significant difference between one viscoelastic or another in regards to postoperative endothelial cell counts. Amvisc Plus works well and we can obtain 0.8 cc of it at a very fair price.

Next a relatively large diameter continuous tear anterior capsulectomy is fashioned (Figures 4 and 5). This can be made with a cystotome or forceps. The optimal size is 5.0 to 6.0 mm in diameter and inside the insertion of the zonules (usually at 7 millimeters). Larger is better than smaller, as there is less subcapsular epithelium and thus lower risk of capsular opacification. Additionally, a larger capsulorhexis makes for an easier cataract operation.

Hydrodissection is then performed utilizing a Pearce hydrodissection cannula on a 3cc syringe filled with BSS. Slow continuous hydrodissection is performed gently lifting the anterior capsular rim until a fluid wave is seen. At this point irrigation is continued until the nucleus tilts on one side, up and out of the capsular bag (Figure 6). If one retracts the capsule at approximately the 7:30 o'clock position with the hydrodissection cannula, usually the nucleus will tilt superiorly. If it tilts in another position, it is simply rotated until it is facing the incision (Figure 7).8

Once the nucleus is tilted some additional viscoelastic can be injected under the nucleus pushing the iris and capsule back. Also, additional viscoelastic can be placed over the nuclear edge to protect the endothelium. The nucleus is emulsified from outside-in while supporting the nucleus in the iris plane with a second instrument, such as a Rhein Medical or Storz Lindstrom Star or Lindstrom Trident nucleus rotator (Figure 8). Once half the nucleus is removed, the remaining one half is tumbled upside-down and approached from the opposite pole (Figure 9). Again, it is supported in the iris plane until the emulsification is completed (Figure 10). Alternatively the nucleus can be rotated and emulsified from the outside edge in, in a carousel or cartwheel type of technique. Finally, in some cases, the nucleus can be continuously emulsified in the iris plane if there is good followability until the entire nucleus is gone.9

This a very fast and very safe technique, and as mentioned before, it is a modification of the iris plane technique taught by Richard Kratz, MD in the late 1970's and 1980's. It is basically “back to Kratz” with help from Brown and Maloney in the modern phacoemulsification, capsulorhexis, hydrodissection and viscoelastic era. Surgery times now range between four and seven minutes with this approach rather than ten to fifteen minutes for endocapsular phacoemulsification. In addition, our capsular tear rate has now gone under 1%. Therefore, we find this technique which to be easier, faster and safer. It is true that in this technique the phacoemulsification tip is closer to the iris margin and also somewhat closer to the corneal endothelium. There is, however, a significantly greater margin of error in regards to the posterior capsule. Care needs to be taken to position the nucleus away from the corneal endothelium and away from the iris margin when utilizing this approach.

If the nucleus does not tilt with simple hydrodissection, it can be tilted with viscoelastic or a second instrument such as a nuclear rotator, Graether collar button or hydrodissection cannula.

The dual function Bausch and Lomb Millennium^TM is excellent for all cataract techniques including “tilt and tumble.” The vacuum is set with a range of 325 to 400 mm Hg and the ultrasound power set in a pulse mode from 10 to 30%. The foot pedal is arranged such that there is surgeon control over ultrasound on the vertical or pitch motion of the foot pedal, and then on the yaw or right motion foot pedal, there will be vacuum control. This allows very efficient emulsification, and the Millennium ^TM is currently our preferred machine. The microflow plus needle with a 30° angle tip works well with the Millennium.

Following completion of nuclear removal, the cortex is removed with the irrigation aspiration hand piece. We prefer a 0.3 mm tip and utilize the universal hand piece with interchangeable tips. A curvilinear tip is used for most cortex removal. Sub-incisional cortex can be aspirated with a Lindstrom right angle sand blasted tip currently manufactured by Rhein and Storz (Figure 11). If there is significant debris or plaque on the posterior capsule, one can attempt some polishing and vacuum cleaning but not so aggressively as to risk capsular tears.

The anterior chamber is reconstituted with viscoelastic and the intraocular lens is inserted utilizing an injector system (Figures 12–13).

Excess viscoelastic is removed with irrigation aspiration. Pushing back on the intraocular lens and slowly turn the irrigation aspiration to the right and left two or three times allows a fairly complete removal of viscoelastic under the intraocular lens.

We favor injection of a miotic and tend to prefer carbachol over miochol at this time, as it is more effective in reducing postoperative intraocular tension spikes and has a longer duration of action. It is best to dilute the carbachol 5 to 1, or one can obtain an excessively small pupil which results in dark vision for the patient at night for one to two days.

At completion of the procedure another drop of antibiotic, steroid and non-steroidal, is placed on the eye. Additionally, one drop of an anti-hypertensive such as Betagan or Alphagan is applied to reduce postoperative intraocular tension spikes.

No patch is routinely utilized for the topical and intracameral approach. If a mini-block of the lids has been performed, this will wear off in thirty to forty-five minutes, and there is usually adequate lid function for a normal blink at the completion of the procedure. Patients are advised that they will have some erythropsia, meaning they will see a pink after image for the rest of the day, but usually this will resolve by the next morning. They are also told that their vision may be a little dark at night from the miotic, and not to be concerned if they wake up at night and their vision seems dimmer.

The patient is seen on the first day postoperative and then at approximately two to three weeks postoperative. At this time a refraction, slit-lamp, and fundoscopic examination is performed. If there is no inflammation, patients are seen again one year postoperative. If at three weeks there is still persistent inflammation, additional postoperative anti-inflammatory medications are recommended, and the patient is asked to return again at two to three months postoperative.

Topical antibiotic, steroid and non-steroidal, are utilized twice a day, usually requiring a 5 cc bottle and three to four weeks of therapy. Occasionally a second bottle of steroid and non- steroidal is necessary if flare and cell persist at the three week examination. There are minimal restrictions, including a request that there be no swimming and no very heavy lifting for two weeks. We consider the ideal postoperative refractive spherical equivalent for a monofocal lens to be −0.62 diopters with less than 0.50 diopters of astigmatism in the same axis as existed preoperatively. Most patients can see 20/30+ and J3+ with this type of correction. 13Monovision can be utilized in the appropriate settings. Good results can also be obtained accommodating or multifocal intraocular lens.

The second eye is done at 2 weeks or greater postoperatively except in rare situations. Any YAG lasers are deferred for 90 days in order to allow the blood aqueous barrier to become intact and capsular fixation to be firm.

We hope other surgeons will find this approach to cataract surgery useful. These techniques must be personalized, and every surgeon will find that slight variations in technique are required to achieve optimum results for their own individual patients in their own individual environment. Continuous efforts at incremental improvement result in meaningful advances in our ability to help the cataract patient obtain rapid, safe, visual recovery following surgery.