TY - JOUR T1 - Calculating the individual probability of successful ocriplasmin treatment in eyes with VMT syndrome: a multivariable prediction model from the EXPORT study JF - British Journal of Ophthalmology JO - Br J Ophthalmol SP - 1092 LP - 1097 DO - 10.1136/bjophthalmol-2017-310874 VL - 102 IS - 8 AU - Christoph Paul AU - Christine Heun AU - Hans-Helge Müller AU - Hans Hoerauf AU - Nicolas Feltgen AU - Joachim Wachtlin AU - Hakan Kaymak AU - Stefan Mennel AU - Michael Janusz Koss AU - Sascha Fauser AU - Mathias M Maier AU - Ricarda G Schumann AU - Simone Mueller AU - Petrus Chang AU - Steffen Schmitz-Valckenberg AU - Sara Kazerounian AU - Peter Szurman AU - Albrecht Lommatzsch AU - Thomas Bertelmann Y1 - 2018/08/01 UR - http://bjo.bmj.com/content/102/8/1092.abstract N2 - Background/Aims To evaluate predictive factors for the treatment success of ocriplasmin and to use these factors to generate a multivariate model to calculate the individual probability of successful treatment.Methods Data were collected in a retrospective, multicentre cohort study. Patients with vitreomacular traction (VMT) syndrome without a full-thickness macular hole were included if they received an intravitreal injection (IVI) of ocriplasmin. Five factors (age, gender, lens status, presence of epiretinal membrane (ERM) formation and horizontal diameter of VMT) were assessed on their association with VMT resolution. A multivariable logistic regression model was employed to further analyse these factors and calculate the individual probability of successful treatment.Results 167 eyes of 167 patients were included. Univariate analysis revealed a significant correlation to VMT resolution for all analysed factors: age (years) (OR 0.9208; 95% CI 0.8845 to 0.9586; p<0.0001), gender (male) (OR 0.480; 95% CI 0.241 to 0.957; p=0.0371), lens status (phakic) (OR 2.042; 95% CI 1.054 to 3.958; p=0.0344), ERM formation (present) (OR 0.384; 95% CI 0.179 to 0.821; p=0.0136) and horizontal VMT diameter (µm) (OR 0.99812; 95% CI 0.99684 to 0.99941, p=0.0042). A significant multivariable logistic regression model was established with age and VMT diameter.Conclusion Known predictive factors for VMT resolution after ocriplasmin IVI were confirmed in our study. We were able to combine them into a formula, ultimately allowing the calculation of an individual probability of treatment success with ocriplasmin in patients with VMT syndrome without FTHM. ER -