Performing phacoemulsification in vitrectomized eyes poses a number of challenges and difficulties—in part because the absence of vitreous tamponade results in a fluctuating anterior chamber. As the nuclear pieces are progressively removed and emulsified, capsular support diminishes and the posterior capsule begins to flap. The problem is accentuated in the presence of a hard nucleus, when there is no epinucleus cushion, possibly resulting in a higher rate of posterior capsular dehiscence.1
When operating on patients with high myopia, we encounter a similar situation. The liquefied vitreous offers little support to the posterior capsule, which tends to flutter, increasing the risk for a posterior capsular rent.2
We have devised a new technique that we call the sandwich technique, which aims to surround the nuclear pieces in an artificial epinucleus, thereby protecting the posterior capsule and permitting in-the-bag phaco. We use Healon5 (Advanced Medical Optics, Inc., Santa Ana, California) to surround the nuclear pieces.
Surgery may be performed either under topical or peribulbar anesthesia, depending on (1) patient comfort and (2) the surgeon's preference. Some patients experience pain once the chamber deepens and the phacoemulsification probe is introduced into the space. In these situations, we supplement anesthesia with intracameral preservative-free xylocaine 2%.
After cleaning and draping, a clear corneal incision is made with a 2.8-mm keratome. We then inject Healon5 and create a sideport incision at the 2-o'clock position using the MVR blade. Next, we inject 0.1 mL of balanced salt solution under the viscoelastic (ie, ultimate soft shell technique3) and complete the capsulorrhexis with a bent 26-gauge needle. A gentle hydrodissection is performed followed by hydrodelineation.
We routinely use the Alcon Universal II phaco machine (Alcon Laboratories, Inc., Fort Worth, Texas) with a 30° tip. The initial power settings depend on the anticipated hardness of the nucleus. We typically begin with an initial flow rate of 25 cc/min and 150 mm Hg vacuum. A low bottle height is preferred, thus avoiding excessive deepening of the anterior chamber. We believe that using a chop technique is safer in these patients because it induces less stress on the bag and the zonules. We prefer the slice-and-separate technique,4 but any chopping method is acceptable.
The phaco probe is buried deep into the nucleus and sliced in half with aide of a sharp chopper. After rotating the nuclear pieces by 90°, the inferior
After emulsifying one or two pieces of the nucleus, the probe is withdrawn from the chamber and Healon5 is injected under the pieces, lifting them and pushing the posterior capsule down. Injecting 0.1 mL of Healon5 over the nuclear pieces ensures stabilization and prevents the pieces from popping out of the bag. The flow rate is then reduced to 18 cc/minute and the vacuum to 100 mm Hg. The phaco probe is then reintroduced into the anterior chamber—it is preferable to bypass continuous irrigation, therefore retaining the viscoelastic for a longer time. The pieces are then impaled and emulsified one by one.
Repeat injections of Healon5, both above and below the remaining pieces, may be necessary to keep the posterior capsule down. Once phacoemulsification is complete, we use the I/A handpiece to remove the cortical matter. Healon5 is injected again, and the IOL is implanted in the capsular bag. We use the two-compartment technique, introduced by Manfred R. Tetz, MD, of Germany,5 to remove Healon5 from the anterior chamber. The sideports and the main incision are then hydrated.
Recently, we used the sandwich technique in a series of 18 patients. Thirteen of these patients previously underwent vitrectomy for nonclearing hemorrhage due to proliferative diabetic retinopathy (n=12) or for the removal of retained intraocular foreign body (n=1); three had traction retinal detachment secondary to Eale's disease; and two had high myopia with posterior staphyloma.
Cataract surgery was uneventful in all patients, with no significant fluctuation of the anterior chamber. Injections of Healon5 were repeated two to four times to maintain stability of the posterior capsule. Postoperative visual acuity (range, 6/12–3/60) was limited by the posterior segment pathology. No patient had a significant rise in the intraocular pressure.
After achieving successful results with the sandwich technique, we have also tested new methods to refine our technique. First, we tried using Viscoat (Alcon Laboratories, Inc.) in a series of 10 patients, maintaining the same machine parameters we used with Healon5. Compared with Viscoat, Healon5 imparted greater stability to the nuclear pieces. Whenever the irrigation was accidentally turned on before impaling the nuclear fragments, we noted that the nuclear pieces went deep into the bag. Second, we experimented with a higher percentage of hydroxypropyl methylcellulose 3% (Visilon GV; Shah and Shah, India) in two patients. Thus far, we have been impressed with the results.
The sandwich technique exploits the superviscous properties of Healon5, which behaves as a cohesive substance at low-flow rates. Due to its high viscosity, Healon5 has the greatest capacity to create and maintain spaces inside the eye; however, at higher flow rates it tends to fracture and behave as a pseudodispersive substance, persisting in the anterior chamber. It is the only substance other than Viscoat to be retained in the anterior chamber at all stages of phacoemulsification.6,7 Flow rates in excess of 25 cc/minute and high vacuum settings (250–300 mm Hg) are needed to remove it from the anterior chamber.8 We therefore use a low flow rate and low vacuum setting, allowing retainment in the anterior chamber during phaco-emulsification. A low flow rate might be expected to compromise followability, but we have not encountered any problems.
The higher viscosity of Healon5 compared with that of Viscoat (200,000 vs 41,000 centipoise, respectively) probably accounts for the greater stability in sandwiching the nuclear fragments. In effect, it cushions the nucleus and separates it from the posterior capsule, which may be valuable for inexperienced surgeons; however, in experienced hands, the two may be equivalent.
We look forward to trying hydroxypropyl methylcellulose 3% in more patients. It may prove to be a viable low-cost alternative.
The unique properties of Healon5 permit retention in the anterior chamber during phacoemulsification, thereby forming a protective cushion around the nucleus. Although we have tried this technique only in vitrectomized eyes and those with high myopia, it could be used in any patient with a floppy posterior capsule.
R. Muralidhar, MD, DNB, FRCS, MRCOphth, is a consultant to Aravind Eye Hospital, Madurai, India. Dr. Muralidhar states that he has no financial interest in the products or companies mentioned. He may be reached at e-mail: firstname.lastname@example.org.
Vikas Sharma, MD, MRCS(Ed), MRCOphth, is a clinical fellow at Moorfields Eye Hospital, London. Dr. Sharma states that he has no financial interest in the products or companies mentioned. He may be reached at tel:+44 7709007350; e-mail: email@example.com.
Rajeev Sudan, MD, is a consultant to the RR Eye Center, New Delhi. Dr. Sudan states that he has no financial interest in the products or companies mentioned. He may be reached at e-mail: firstname.lastname@example.org.