June 2020
Volume 61, Issue 7
Free
ARVO Annual Meeting Abstract  |   June 2020
Mathematical models using OCT parameters to predict macular hole closure rates: a post-hoc analysis from a prospective clinical trial
Author Affiliations & Notes
  • Rodrigo Jorge
    Ophthalmology, Ribeirao Preto Medical School, University of Sao Paulo, Ribeirao Preto, Sao Paulo, Brazil
  • Tomas Teixeira Pinto
    Ophthalmology, Ribeirao Preto Medical School, University of Sao Paulo, Ribeirao Preto, Sao Paulo, Brazil
  • Karlos Italo Viana
    Ophthalmology, Ribeirao Preto Medical School, University of Sao Paulo, Ribeirao Preto, Sao Paulo, Brazil
  • Andre Messias
    Ophthalmology, Ribeirao Preto Medical School, University of Sao Paulo, Ribeirao Preto, Sao Paulo, Brazil
  • Footnotes
    Commercial Relationships   Rodrigo Jorge, None; Tomas Pinto, None; Karlos Viana, None; Andre Messias, None
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science June 2020, Vol.61, 4396. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Rodrigo Jorge, Tomas Teixeira Pinto, Karlos Italo Viana, Andre Messias; Mathematical models using OCT parameters to predict macular hole closure rates: a post-hoc analysis from a prospective clinical trial. Invest. Ophthalmol. Vis. Sci. 2020;61(7):4396.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose : To study the role of preoperative spectral-domain optic coherence tomography (OCT) parameters to predict macular hole (MH) closure after pars-plana vitrectomy (PPV) with internal limiting membrane (ILM) peeling

Methods : Files from 65 patients (65 eyes) with symptomatic MH (stage II or higher) and no significant cataract that were included in a randomized clinical trial were analyzed. Briefly, patients were randomized to either combined phacoemulsification and PPV (PHACO-PPV) or PPV and deferred PHACO (PPV-PHACO) and followed for 48 weeks. ILM peeling was performed in a 360-degree fashion around all MH after brilliant blue staining. Preoperative OCT scans (Spectralis, Heidelberg Engineering) were analyzed for measurements of MH height, minimum linear and base diameters, and to determine the macular hole index (MHI). The leverage of these measurements on postoperative MH closure was determined by multivariate nominal logistic model. Additionally, preoperative best-corrected visual acuity, duration of symptoms, age, and gender were also considered as potential influencing features.

Results : Considering congregated data from both groups, mean (± SE) preoperative BCVA (logMAR) was 0.91 ± 0.04 and improved significantly to 0.59 ± 0.05 (P<0.001) after 48 weeks. 20, 25 and 20 patients had macular hole stage II, III and IV, respectively, according to Gass classification. MH closure was observed on 49 out of 65 cases (75%). Mean preoperative OCT parameters values were: MHI: 2.36 ± 0.61; height = 444.5±7.7 µm; base-diameter = 1057.6±41.1 µm; and minimum linear diameter = 475.0±25.1 µm. There was a significant influence of MH base-diameter and MHI on MH closure (P = 0.0005). We estimated a high probability of MH closure for patients with preoperative OCTs showing base diameter smaller than 884.2 µm ± 39.0 µm, and this likelihood decreases from this point at ratio of 11.16 % ± 0.84 % / (100 µm) (Figure 1). We also estimated a high probability of MH closure for patients with preoperative OCTs presenting a MHI smaller than 2.3 (Figure 2).

Conclusions : It is possible to estimate the probability of macular closure using pre-operative OCT images. These estimates could be used to improve patient and surgeon awareness of anatomical success likelihood and influence the choice of surgical technique. Data from larger cohorts should be used to confirm these results.

This is a 2020 ARVO Annual Meeting abstract.

 

 

×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×