Wound construction definitely can affect the rate of postoperative hypotony. Oblique cannula insertion re- duces the rate of bleb formation and hypotony compared with perpendicular cannula insertion in 25-gauge pars plana vitrectomy eyes. 4 India ink, anterior segment spec- tral optical coherence tomography, and spectrophotometry studies on human cadaveric eyes also demonstrated better wound integrity with oblique incisions compared with perpendicular incisions. 5,6 Spectrophotometric analysis of vitreous aspirates demonstrated that 23-gauge sclerotomies leaked more than 25-gauge sclerotomies. Thus, as Byeon and associates have correctly observed, there are many factors that need to be addressed when performing microincision surgery. Wound construction, patient characteristics, and surgical maneuvers all play an important role in surgical outcome and prevention of complications. We thank the authors for their comments. OMESH P. GUPTA ALLEN C. HO Philadelphia, Pennsylvania PETER K. KAISER Cleveland, Ohio CARL D. REGILLO Philadelphia, Pennsylvania SANFORD CHEN Laguna Hills, California DAVID S. DYER Kansas City, Missouri PRAVIN U. DUGEL Phoenix, Arizona SUNIL GUPTA Pensacola, Florida JOHN S. POLLACK Chicago, Illinois REFERENCES 1. Byeon SH, Lew YJ, Kim M, Kwon OW. Wound leakage and hypotony after 25-gauge sutureless vitrectomy: factors affect- ing postoperative intraocular pressure. Ophthalmic Surg La- sers Imaging 2008;39:94 –99. 2. Gupta OP, Weichel ED, Regillo CD, et al. Postoperative complications associated with 25-gauge pars plana vitrectomy. Ophthalmic Surg Lasers Imaging 2007;38:270 –275. 3. Kunimoto DY, Kaiser RS. Incidence of endophthalmitis after 20- and 25-gauge vitrectomy. Ophthalmology 2007;114:2133– 2137. 4. Hsu J, Chen E, Gupta OP, et al. Post-operative hypotony after 25-gauge sutureless vitrectomy using oblique versus direct cannula insertions. Retina 2008;28:937–940. 5. Gupta OP, Maguire JI, Eagle RC, et al. The competency of pars plana vitrectomy incisions: a comparative histologic and spectrophotometric analysis. Am J Ophthalmol. Forthcoming. 6. Taban M, Ventura A, Sharma S, Kaiser PK. Dynamic evalu- ation of sutureless vitrectomy wounds: an optical coherence tomography and histopathology study. Ophthalmology. Forthcoming. Ocular Adnexal Lymphoid Tumors: Progress in Need of Clarification EDITOR: WITH NUMEROUS UNANSWERED QUESTIONS ABOUT OCU- lar adnexal lymphomas (OALs), Jakobiec’s recent perspec- tive is most welcome. 1 He highlighted the need for identifying prognostic factors, lamenting the lack of any incontestable pathologic or immunophenotypic criteria and remarking that clinical staging transcended tissue studies. Jackobiec also observed how lymphoproliferations previously diagnosed as benign or atypical lymphoid hyperpla- sia are now regarded as extranodal marginal zone lympho- mas (EMZL). Finally, Jakobiec stressed the need for multicenter and multidisciplinary studies on OAL using World Health Organization (WHO) guidelines. He praised a large study by Ferry and associates, 2 acknowledg- ing a few shortcomings, particularly lack of long-term follow-up. I wish to endorse Jakobiec’s views and make a few additional points. In 2004, my associates and I reported a cohort of 261 consecutive patients, 136 of whom had EMZL. 3 As recom- mended by Jakobiec, we subtyped the primary OAL ac- cording to the WHO Lymphoma Classification and analyzed risk factors for local recurrence, systemic disease development, and death from lymphoma, which occurred in 63, 62, and 57 patients, respectively. A Ki-67 tumor cell growth fraction 10% correlated with a reduction in the 10-year survival from 80% to 50% (log-rank test 0.001). Others 4 have also demonstrated the prognostic significance of Ki-67 (not investigated by Ferry and asso- ciates). I, therefore, recommend that this prognostic indi- cator be included in future OAL studies. I concur with Jakobiec about the prognostic value of clinical staging. Readers will be interested to know that a novel staging system for primary OAL has recently been developed by myself and colleagues in the ophthalmic division of the Tumor Node Metastasis (TNM) Commit- tee of the American Joint Committee on Cancer (AJCC). 5 This is the first staging system specifically devised for OAL. Briefly, OAL are TNM-staged as: “I” if confined to conjunctiva; “II” with orbital involvement; “III” with cutaneous involvement; and “IV” if there is direct extraorbital spread. This system overcomes limita- tions of the Ann Arbor system, which does not differen- tiate “high-risk” from “low-risk” OALs. With regard to atypical and reactive lymphoid hyperplasia as well as to lymphoplasmocytic lymphoma, these were diagnosed in only 2%, 11%, and 4%, respectively, of the patients reported in our 2004 study, in keeping with Jakobiec’s observation about terminology. There has been a shift in diagnostic patterns, with previously-thought- benign ocular adnexal lymphoproliferative lesions being reclassified as EMZL. Interestingly, the same trend has CORRESPONDENCE VOL. 146,NO. 5 791