Biometry Prediction Error of the Postoperative Refraction After Combined Pars Plana Vitrectomy and Cataract Removal with Intraocular Lens Implantation: Phacoemulsification Versus Pars Plana Lensectomy

Abstract

Abstract: : Purpose: To investigate the range of biometry prediction error (BPE) of the postoperative refraction after combined pars plana vitrectomy (PPV) and cataract removal via phacoemulsification or pars plana lensectomy and intraocular lens (IOL) implantation, and to investigate factors associated with a postoperative prediction error >1.00 diopter (D) spherical equivalent. Methods: The medical records were reviewed of all patients who underwent combined PPV and cataract removal with IOL implantation at a single institution between January 1, 2000 and December 31, 2000. The BPE was calculated as the SRK-T predicted refraction minus the postoperative spherical equivalent. It was also computed using the Holladay formula. Results: The study included 74 eyes of 74 patients; 47 patients who underwent combined pars plana lensectomy with PPV and IOL implantation (PPL/PPV group), and 27 patients who underwent combined phacoemulsification with PPV and IOL implantation (Phaco/PPV group). The mean age (standard deviation) was 60 (15) years; 61 (17) years in PPL/PPV group, and 59 (12) years in the Phaco/PPV group (P=0.48). The PPL/PPV group had a greater BPE 0.7 (1.5) D than the Phaco/PPV group -0.05 (1.6) D (P=0.05). (The results were the same using both the SRK-T and Holladay formulas) The number (%) of patients with a BPE >1.00 D was 37/74 (50%); 26/47 (55%) in the PPL/PPV group and 11/27 (41%) in the Phaco/PPV group (P=0.33) using the SRK-T formula. Results were similar using the Holladay formula, 27/47 (57%) for the PPL/PPV and 11/27 (41%) for the Phaco/PPV (P=0.25); 38/57 (51%) overall. Risk factors significantly associated with a BPE >1.00 D included female gender and previous incisional ocular surgery. Sixty-six percent (21/32) of the females versus 38% (16/42) of the males had a BPE of >1.00 D using both SRK-T (P=0.035) and Holladay (P=0.017) formulas. Seventy-four percent (14/19) of patients with previous surgery had a BPE of >1.00 D versus 42% (23/55) of those without previous surgery (P=0.033) using the SRK-T formula. Results were similar using the Holladay formula (P=0.046). Conclusions: A significant proportion (50%) of patients undergoing cataract removal with concurrent PPV have a biometry prediction error > 1.00 D. The PPL/PPV group has a greater biometry prediction error than the Phaco/PPV group. Other statistically significant risk factors include female gender and previous incisional ocular surgery.

Keywords: treatment outcomes of cataract surgery • refractive surgery: phakic IOL • cataract