Calculating the individual probability of successful ocriplasmin treatment in eyes with VMT syndrome: a multivariable prediction model from the EXPORT study

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...

Full description

Saved in:
Bibliographic Details
Main Authors: Paul, Christoph (Author) , Koss, Michael Janusz (Author)
Format: Article (Journal)
Language:English
Published: 2018
In: British journal of ophthalmology
Year: 2017, Volume: 102, Issue: 8, Pages: 1092–1097
ISSN:1468-2079
DOI:10.1136/bjophthalmol-2017-310874
Online Access:Verlag, kostenfrei, Volltext: http://dx.doi.org/10.1136/bjophthalmol-2017-310874
Verlag, kostenfrei, Volltext: http://bjo.bmj.com/lookup/doi/10.1136/bjophthalmol-2017-310874
Get full text
Author Notes:Christoph Paul, Christine Heun, Hans-Helge Müller, Hans Hoerauf, Nicolas Feltgen, Joachim Wachtlin, Hakan Kaymak, Stefan Mennel, Michael Janusz Koss, Sascha Fauser, Mathias M Maier, Ricarda G Schumann, Simone Mueller, Petrus Chang, Steffen Schmitz-Valckenberg, Sara Kazerounian, Peter Szurman, Albrecht Lommatzsch, Thomas Bertelmann
Description
Summary: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.
Item Description:Published online first 31 October 2017
Gesehen am 28.06.2018
Physical Description:Online Resource
ISSN:1468-2079
DOI:10.1136/bjophthalmol-2017-310874