Session Title: Vitreo Retinal Surgery I
Session Date/Time: Thursday 11/09/2014 | 08:00-10:00
Paper Time: 08:40
First Author: : M.Dogramaci UK
Co Author(s): : T.H. Williamson M.K. Bayazit A. Cunningham
We recently demonstrated that symptomatic retinal shifts are common in eyes following PPV for RRD. Because of the alarming symptoms of the retinal shifts, it is very difficult to investigate the influencing factors through large scale studies. Therefore we used SRE technology to create computer models to assist in studying the influencing factors. In this second paper we demonstrate the methods of generating a reliable computer model through RSE using data presented in our first paper.
University Hospital of North Durham, St. Thomas' Hospital, Imperial College
The Clinical data used in SRE were given in our previous paper. The retina was modeled as a sphere with a diameter of 24mm and thickness of 0.2 mm. The Subretinal fluid (SRF) was modeled as an elevated area at the posterior pole with a radius of 5-10mm and a height of 0.5-5mm. The optic disc was 1.5mm in diameter and attached to the underlying choroido-scleral complex. The hydrostatic pressure of the SRF, the buoyancy pressure of the gas tamponade and the gravity acceleration were all applied with their vector directions depended on the type of the posture. Young’s modulus of the retina was 0.03 mega Pascal (MPa). Laterality was defined by the location of the optic disc. Attached retina was assigned a bond contact while the detached retina was assigned Friday 12 Septemberctionless contact to the underlying structure. The values for the sample shift points were determined using directional deformation results and correlated to the values of the sample points detected on the AF images.
A total of 15 simulations for eleven eyes were sufficient to establish statistically significant correlation between the sample shift points recorded in the computer models with 151 sample points detected on the AF images. For 6 eyes, that were instructed to assume an initial temporal posture, two simulations were necessary for each eye to achieve statistically significant correlations for both vertical and horizontal sample shift points. The first simulation was for an initial temporal posture followed by a second simulation for the intermediate upright posture (when the patients were stood up in their way home). For 2 eyes that had no initial posture instructions and assumed upright posture immediately after the operation, and for 3 eyes with an initial posture instruction of face up, only the intermediate upright posture simulation was necessary to achieve statistically significant correlation. The mean SRF density, height and diameter used to simulate the initial postures in 8 simulations were 2500 Kg/mm3, 1,53 mm and 8.66 mm, this is compared to 9888.88 Kg/mm3, 1,50 mm and 10.77 mm values used in 7 simulations for intermediate posture. The difference between the densities were statistically significant p=0.008
The model showed that retinal shifts are influenced by the initial two hour posturing as well as the intermediate upright posture that follows when the patients are discharged. The ultimate long term posture resumed by the patient at home did not seem to influence the shift. The SRF becomes higher in density and distributes over a wider area after 2 hours from the initial posturing. Therefore longer duration of initial posturing may reduce the risk of significant retinal shifts. Additionally understanding data processing and information flows of SRE technology could help in the application of this technology in the field of viteoretinal surgery and designing reliable models to study surgical complications on larger scales.