October 2011




Chief medical editor's corner of the world

Part I: Complication rates of phaco vs. manual ECCE among initial surgical trainees at UCSF

by Ayman Naseri, M.D.


Last year, I received an urgent phone call from a Bay Area surgeon who had paused during cataract surgery on a pseudoexfoliation patient when it became apparent that the zonules were extremely deficient. The lens was so mobile that it was hard to complete the capsulorhexis and the nucleus was very brunescent. With the patient still on the table, my advice was to convert to a manual ECCE. However, the surgeon volunteered that she had never performed or been trained in manual ECCE and as a result preferred to abort the operation and refer the patient to me. The question of whether manual ECCE could become a vanishing art is an important one as increasing numbers of residency programs reduce or eliminate this component of cataract surgical training. I asked our UCSF residency program director, Ayman Naseri, M.D., to summarize his recent study regarding complication rates of phaco vs. manual ECCE among initial surgical trainees at UCSF. The study was published in the June issue of Ophthalmology. Ayman was an unusually precocious and gifted surgeon as a UCSF resident, where I was one of his early phaco attendings. Now that he oversees cataract surgical training for UCSF residents, it is wonderful to see him apply his teaching skills and outstanding analytical insights to this important and often challenging endeavor.

I then asked Rosa Braga-Mele, M.D., Devin Gattey, M.D., Lisa Park, M.D., and Jack Dodick, M.D., to comment on the study and the implications. All four are noted experts in training residents in cataract surgery and shared their insights and experiences from the University of Toronto, Oregon Health & Science University, and NYU School of Medicine (Manhattan Eye and Ear Infirmary), respectively.

David F. Chang, M.D., chief medical editor


Ayman Naseri, M.D., UCSF residency program director

All ophthalmology residency programs in the United States must teach cataract surgery in order to maintain accreditation, but each program makes its own choices as to how to best train young surgeons. Although most training is focused on teaching phacoemulsification techniques, some believe that extracapsular cataract extraction (ECCE) is an important part of surgical education in cataract removal. Some educators argue that ECCE experience imparts basic surgical skills such as suturing and wound construction that are essential to the beginning ophthalmology resident.1 Others contend that ECCE is the standard "fall back" technique in the event of certain phacoemulsification complications and therefore should be learned prior to phacoemulsification.2 But many believe that ECCE is an outdated method and that training residents in ECCE prior to phacoemulsification is unnecessary.3 Still others argue that ECCE should be taught, but not as an introductory technique for beginning surgeons. A report at the 2009 meeting of the Association of University Professors of Ophthalmology showed that in 2002, 73% of U.S. ophthalmology residents began learning cataract extraction with ECCE,4 whereas by 2009 only 55% of residents began with ECCE (Ta, Chris. Do Residents Still Perform ECCE? Lecture presented at: Association of University Professors of Ophthalmology, 43rd Annual Meeting, Jan 29, 2009; Indian Wells, Calif.). This trend suggests a shift away from teaching ECCE to ophthalmology residents in the U.S. At UCSF, we continue to train our first-year residents in ECCE prior to teaching phacoemulsification. Through reviewing surgical complications at our weekly morbidity and mortality conference, we realized that we needed to ask a more fundamental question: Are we putting patients at risk for complications by using one technique over another? Few studies have compared outcomes of the two surgical techniques at the earliest stage in resident experience, when the surgeon learning curve is steepest. We recognized that, because of pre-op differences in case selection between ECCE and phacoemulsification, a true comparison is difficult in a retrospective study. For example, patients with denser cataracts might be more likely to be scheduled for ECCE because of the difficulty of phacoemulsification in these patients. Another potential difficulty is that temporally in each resident's experience, the ECCE surgeries occurred before the phacoemulsifications, as is standard practice in the UCSF residency program, thus imparting an experiential advantage to the phacoemulsification cases. Finally, since the surgeries occurred in three different hospitals, one or more of the sites may have been more likely to employ a certain procedure based on the pathology inherent to its patient population. Nonetheless, we believed it was important to have some outcomes data to compare the two procedures from a patient safety perspective. We retrospectively reviewed first-year ECCEs and the first 10 phacoemulsifications performed by the same residents between 2002 and 2008, looking specifically at vitreous loss, posterior chamber intraocular lens (PCIOL) placement, and reoperation within 90 days of the initial surgery. Because ECCEs were performed prior to phacoemulsifications and because ECCE patients often have dense cataracts, we hypothesized that the rate of complications would be higher in ECCE cases.5 We identified 171 eyes of 160 patients who underwent ECCE and 255 eyes of 242 patients who underwent phacoemulsification. Vitreous loss occurred in 29 (17.0%) of 171 eyes that underwent ECCE and in 43 (16.9%) of 255 eyes that underwent phacoemulsification (P=0.95). A PCIOL was initially placed in 153 (89.5%) ECCE cases as compared to 248 (97.3%) phacoemulsification cases (P=0.002). Reoperation within 90 days of cataract surgery took place in 5 (2.9%) ECCE cases and in 13 (5.1%) phacoemulsification cases (P=0.31). When pre-op BCVA, hospital, and age were controlled for in a multivariate analysis, there was no evidence of a relationship between type of cataract extraction and vitreous loss rate (P=0.86), placement of a PCIOL (P=0.10), or reoperation rate (P=0.10).

Although the vitreous loss rates in our study were uncomfortably high in both groups, we felt reassured that there was no ethical dilemma posed by training our residents in ECCE when comparing outcomes between the two groups. We had already studied our complication rate for resident-performed cataract surgery over the course of our 3-year program and found the vitreous loss rate to be 3.1%. But because of the high rates we found in this study, we have been working diligently to decrease the complication rate for both techniques. Other issues that we have identified in the course of looking at teaching ECCE in our residency program include the adoption of small incision ECCE techniques and the decreasing numbers of experienced ECCE educators over time. Many of our current teaching faculty trained in the "phaco" era and are therefore less comfortable teaching ECCE. At the same time, we are impressed by the advances in small incision ECCE popularized by our international colleagues, and we have begun to train our residents in these methods as well.


1. Smith JH. Teaching phacoemulsification in US ophthalmology residencies: can the quality be maintained? Curr Opin Ophthalmol 2005;16:27-32.

2. Stewart JM. Phacoemulsification performed by residents [letter]. J Cataract Refract Surg 2007;33:755.

3. Bhagat N, Nissirios N, Potdevin L, et al. Complications in resident-performed phacoemulsification cataract surgery at New Jersey Medical School. Br J Ophthalmol 2007;91:1315-7.

4. Rowden A, Krishna R. Resident cataract surgical training in United States residency programs. J Cataract Refract Surg 2002;28:2202-5.

5. De Niro J, Biebesheimer J, Porco TC, Naseri A. Early resident-performed cataract surgery. Ophthalmology. 2011 Jun;118(6):1215-1215.

Part I: Complication rates of phaco vs. manual ECCE among initial surgical trainees at UCSF Part I: Complication rates of phaco vs. manual ECCE among initial surgical trainees at UCSF
Ophthalmology News - EyeWorld Magazine
283 110
283 110
True, 10