Secondary Logo

Journal Logo


Topical–intracameral anesthesia without preoperative mydriatic agents for Descemet-stripping automated endothelial keratoplasty and phacoemulsification cataract surgery with intraocular lens implantation

Oberg, Thomas J. MD; Sikder, Shameema MD; Jorgensen, Adam J. BS; Mifflin, Mark D. MD

Author Information
Journal of Cataract & Refractive Surgery: March 2012 - Volume 38 - Issue 3 - p 384-386
doi: 10.1016/j.jcrs.2011.12.025
  • Free


In recent years, the use of Descemet-stripping automated endothelial keratoplasty (DSAEK) and its variants have largely supplanted penetrating keratoplasty (PKP) for the treatment of Fuchs corneal endothelial dystrophy and other causes of endothelial failure. The traditional triple procedure combined PKP with phacoemulsification or extracapsular cataract extraction and intraocular lens (IOL) implantation. This older approach necessitated complete ocular anesthesia and akinesia as well as pharmacologic dilation preoperatively. A new triple procedure comprising phacoemulsification and IOL implantation followed by an earlier version of endothelial keratoplasty was described by Terry and Ousley in 2003.1 In recent years, improvements in endothelial keratoplasty, particularly the DSAEK technique, have allowed smaller incisions and shorter surgery times, enabling the use of topical anesthesia. We describe a modification of the triple procedure that uses only topical and intracameral anesthetic agents for pain control and only intracameral lidocaine 1% (Xylocaine-MPF) for dilation.


No dilating drops are used. Intravenous sedation is used as necessary, typically midazolam and/or fentanyl, to maintain a light anxiolytic state without affecting patient responsiveness. The topical and systemic medication regimens are similar to those used for patients having cataract surgery at our institution.

Donor tissue is prepared in the operating room using a Moria CB keratome and artificial anterior chamber (Moria). A 300 or 350 μm cutting head is selected based on precut pachymetry. An ophthalmic viscosurgical device (OVD) (sodium hyaluronate [Healon]) is placed on the endothelium before the endothelium is mounted on the artificial anterior chamber system, which is pressurized with corneal storage media (Optisol GS, Bausch & Lomb). After the tissue is cut, the system is carefully disassembled and the tissue transferred atraumatically back into the storage medium.

Mydriasis and additional anesthesia are attained by injection of 0.3 mL of lidocaine 1% without epinephrine through a 1.0 mm stab incision (Video, available at Additional topical lidocaine 1% drops are administered several times during the procedure.

The OVD is used to inflate the anterior chamber and enhance pupil dilation, and a 4.5 mm groove is created at the temporal limbus using a guarded diamond blade set at 380 μm. A 2.8 mm clear corneal incision is made at the base of the groove, followed by capsulorhexis, hydrodissection, and/or hydrodelineation. The lens nucleus is phacoemulsified and/or aspirated using a modified chop technique followed by cortical cleanup. The bag is then reinflated with OVD, and the IOL is inserted and positioned. Two additional 1.0 mm stab incisions are placed for graft insertion and positioning. Descemet membrane is stripped from the host as described by Melles et al.,2 followed by 4 paracentral venting incisions as popularized by Price and Price.3 After removal of the OVD with irrigation and aspiration, an anterior chamber maintainer is placed into the superior paracentesis and the incision enlarged to its full 4.5 mm width. The donor tissue is punched to the appropriate diameter, transferred carefully to a Busin glide (Moria Surgical) and pulled into the anterior chamber using a 25-gauge Grieshaber inner limiting membrane retinal forceps (Alcon Laboratories, Inc.) inserted through a superonasal stab incision.

The anterior chamber is filled with air injected via a 30-gauge cannula and syringe. Interface fluid is released as necessary through the venting incisions and final graft position obtained with a microroller. The attempted endpoint is close to 100% air fill of the anterior chamber and firm pressurization for several minutes, followed by a release of air to maintain a 100% bubble but physiologic pressure. Cyclopentolate 1% is instilled following final manipulation of the air bubble. A collagen shield soaked in gatifloxacin and dexamethasone is then placed and the eye lightly patched. Patients remain supine for 1 hour in the recovery room and are then examined upright at the slitlamp, with the desired endpoint being an air–fluid meniscus above the lower edge of the dilated pupil. If necessary, air is released using sterile technique and patients are discharged with instructions to remain supine as much as possible until seen the following day.


A retrospective case review of 32 consecutive patients at our institution who had combined phacoemulsification, IOL implantation, and DSAEK for Fuchs endothelial dystrophy was performed. Suitability for topical anesthesia was determined by the surgeon during the preoperative assessment and informed consent process, with most patients receiving topical anesthesia. The IOL power was targeted to approximately −1.0 diopter to counteract the expected hyperopic shift seen with DSAEK grafts in our experience. Three drops of bupivacaine 0.75% and 1.0 cm of lidocaine jelly 2.0% were applied in the preoperative area.

The mean procedure time in the patients was 69 minutes ± 11 (SD). No patient reported significant intraoperative or postoperative pain, no grafts dislocated or required bubbling or repositioning, and there were no primary graft failures. There were no cases of pupillary block even though no patient received a peripheral iridectomy. The mean logMAR corrected distance visual acuity improved from 0.40 ± 0.18 preoperatively to 0.26 ± 0.16 at 6 months. The mean endothelial cell loss for 11 available patients was 31% with a mean follow-up of 9 months.


Topical anesthesia for endothelial keratoplasty, described by Fang and Hamill4 in a small series of 7 eyes, has many known advantages that have been well noted in the cataract literature.5,6 The triple procedure combining endothelial keratoplasty and phacoemulsification and IOL implantation offers multiple advantages over a staged procedure in the correct setting, including reduced cost and faster visual rehabilitation.1,7,8 A combination of these techniques, namely the new triple procedure and topical anesthesia, yielded excellent outcomes in our retrospective series of patients.

Although the use of intracameral lidocaine 1% without epinephrine has been reported in both phacoemulsification plus IOL implantation and DSAEK for dilation and pain control,4,9 we believe this method is particularly well suited to the new triple procedure. The half-life of lidocaine in the iris is 8 to 9 minutes,10 and in our experience, the pupil begins to constrict when the OVD is removed from the anterior chamber after Descemet stripping. This is helpful for 3 reasons when transitioning from IOL implantation to DSAEK. First, dilation may cause increased risk for trauma to the donor endothelium because of the exposed surface of the IOL during graft insertion, manipulation, or positioning. Second, wide dilation might allow the IOL to decenter or the optic to prolapse out of the capsular bag during anterior chamber manipulation and/or pressurization with air or balanced salt solution after graft insertion. Third, when the new triple procedure is performed using preoperative topical dilating drops or dilute epinephrine added to intracameral lidocaine 1%, intracameral miotics must often be injected to accomplish pupil constriction for insertion of the donor button. This may interfere with postoperative dilation, which is helpful in preventing pupillary block. Many surgeons perform an inferiorly located prophylactic peripheral iridectomy; however, we have avoided this step because of concerns about potential bleeding or unwanted optical side effects. In our experience, careful air-bubble management, postoperative dilation, and patient positioning have been sufficient to avoid pupillary block. Insufficient dilation to allow safe phacoemulsification and IOL implantation was not encountered in our series but should be considered as a potential pitfall of this technique. Should this occur, we suggest the use of additional intracameral lidocaine 1% with epinephrine or iris retention devices to achieve safe visualization.

Patient cooperation is necessary to perform safe, successful DSAEK, phacoemulsification, and IOL implantation using topical–intracameral anesthesia. Increased eye movement necessitates continual communication between surgeon and patient to prevent complications. Regional injection anesthesia may be more appropriate for selected patients or for surgeons early in the learning curve. Even so, we have found the advantages to outweigh potential concerns. Avoidance of injection anesthesia completely eliminates the risk for globe perforation, retrobulbar hemorrhage, optic nerve injury, and potential cardiac or respiratory events associated with the block itself or sedation required to administer the block. The ability of patients to direct their gaze during manipulation of the DSAEK tissue over the air bubble may be helpful for efficient graft positioning and seating. The omission of preoperative dilating drops, the short-acting dilation of intracameral lidocaine, and the elimination of intraoperative miotics allow more effective pupil management and reduction of unnecessary medications. Finally, this technique has been well-received by patients and surgery-center staff. In our center, using similar anesthesia protocols for DSAEK, phacoemulsification and IOL implantation, and phacoemulsification and IOL implantation alone have resulted in better efficiency, shorter operating-room times, and more consistency in postoperative care and instruction.

In summary, the use of lidocaine 1% without epinephrine for dilation and anesthesia for the new triple procedure (DSAEK, phacoemulsification, and IOL implantation) was effective in our series and not associated with complications. We believe this technique offers a viable alternative to more traditional approaches and may result in better safety and efficiency. Careful patient selection and surgeon learning curve must be considered when evaluating or adopting new techniques.


1. Terry MA, Ousley PJ. In pursuit of emmetropia: spherical equivalent refraction results with deep lamellar endothelial keratoplasty (DLEK). Cornea. 2003;22:619-626.
2. Melles GRJ, Wijdh RHJ, Nieuwendaal CP. A technique to excise the Descemet membrane from a recipient cornea (descemetorhexis). Cornea. 2004;23:286-288.
3. Price FW Jr, Price MO. Descemet’s stripping with endothelial keratoplasty in 200 eyes; early challenges and techniques to enhance donor adherence. J Cataract Refract Surg. 2006;32:411-418.
4. Fang JP, Hamill MB. Descemet’s stripping endothelial keratoplasty under topical anesthesia. J Cataract Refract Surg. 2007;33:187-188.
5. Fichman RA. Use of topical anesthesia alone in cataract surgery. J Cataract Refract Surg. 1996;22:612-614.
6. Tsoumani AT, Asproudis IC, Damigos D. Tetracaine 0.5% eyedrops with or without lidocaine 2% gel in topical anesthesia for cataract surgery. Clin Ophthalmol. 7. 2010. 967-970. Available at: Accessed November 11, 2011.
7. Covert DJ, Koenig SB. New triple procedure: Descemet’s stripping and automated endothelial keratoplasty combined with phacoemulsification and intraocular lens implantation. Ophthalmology. 2007;114:1272-1277.
8. Terry MA, Shamie N, Chen ES, Phillips PM, Shah AK, Hoar KL, Friend DJ. Endothelial keratoplasty for Fuchs’ dystrophy with cataract; complications and clinical results with the new triple procedure. Ophthalmology. 2009;116:631-639.
9. Nikeghbali A, Falavarjani KG, Kheirkhah A, Kheirkhah A, Bakhtiari P, Kashkouli MB. Pupil dilation with intracameral lidocaine during phacoemulsification. J Cataract Refract Surg. 33. 2007. 101-103. Available at: Accessed November 11, 2011.
10. Anderson NJ, Woods WD, Kim T, Rudnick DE, Edelhauser HF. Intracameral anesthesia: in vitro iris and corneal uptake and washout of 1% lidocaine hydrochloride. Arch Ophthalmol. 117. 1999. 225-232. Available at: Accessed November 11, 2011.


Video. The video demonstrates combined phacoemulsification, IOL implantation, and DSAEK using topical and intracameral anesthesia. Injection of lidocaine 1% results in adequate dilation, which is augmented by the addition of an OVD. Iris dilation is maintained during cataract surgery, including capsulorhexis. Following phacoemulsification and IOL implantation, iris constriction occurs with irrigation and aspiration of the OVD from the anterior chamber. The DSAEK donor tissue is inserted into the anterior chamber with a mid-sized iris covering the lens and capsulorhexis. The anterior chamber is filled with air with a smaller pupil, permitting movement of the DSAEK donor tissue into final position.
© 2012 by Lippincott Williams & Wilkins, Inc.