We’re interested in the article by Mo and colleagues[1] about“Evaluation of hyperreflective foci as a prognostic factor of visual outcome in retinal vein occlusion”. They performed a multiple linear regression analysis with backward elimination to analyse the association of best corrected visual acuity (BCVA) with the numbers of hyperre flective foci (HF)in different layers. In addition, they summarized the results of multivariate regression analyses in Table 3[1].
However, it seems that Table 3[1] is not consisting of the results of a multiple regression analyses, rather, it is showing the results of Pearson correlation analysis. Surely, Pearson correlation analysis is already being done in multiple regression analyses automatically, but the main results of the regression analysis must have included the beta coefficients and related P values. And, Table 3[1] is lacking of those values.As it is already known, the multiple linear regression can be used not only to identify the association between two variables,but also help to forecast impacts of changes and predict trends.For instance, a multiple linear regression can tell you how much BCVA is expected to increase (or decrease) for every a number increase (or decrease) in HF.
We believe that, beta coefficients and related P values will also be crucial in understanding the exact association between BCVA and the numbers of HF in different layers.
Conflicts of Interest: Bayat AH, None; Çakır A, None;Erden B, None; Bölükbaşı S, None; Şehirli ŞG, None.
REFERENCE
1 Mo B, Zhou HY, Jiao X, Zhang F. Evaluation of hypereflective foci as a prognostic factor of visual outcome in retinal vein occlusion. Int JOphthalmol 2017;10(4):605-612.
Author Reply to the Editor
Dear Editor,
Thanks for Dr. Alper Halil Bayat et al’s comments on our manuscript entitled “Evaluation of hyperreflective foci as a prognostic factor of visual outcome in retinal vein occlusion”.A little pity that Ididn’t put the beta coefficients and related P values in this paper, which may be confusing to the readers.Here is the table including the beta coefficients and related P values (Table 1).
Table 1 Results of multivariate regression on numbers of HF as a prognostic parameter for poor visual outcome
CRVO: Central retinal vein occlusion; BRVO: Branch retinal vein occlusion; ORL: Outer retinal layer; IRL: Inner retinal layer; RATIO:The ratio of HF numbers to the total areas of vitreous cavity.
Retinal layer 0.025 0.931 0.116 0.618 0.066 0.817 ORL -0.564 0.026 -0.675 0.000 -0.716 0.013 IRL 0.332 0.165 0.475 0.057 0.360 0.181 RATIO -0.482 0.052 -0.007 0.976 0.053 0.825
Bin Mo, Hai-Ying Zhou, Xuan Jiao, Feng Zhang
Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing 100730, China
Received:2017-08-05
Accepted:2017-12-15
DOl:10.18240/ijo.2018.05.30
Citation: Bayat AH, Çakır A, Erden B, Bölükbaşı S, Şehirli ŞG.Comment on “Evaluation of hyperreflective foci as a prognostic factor of visual outcome in retinal vein occlusion”. Int J Ophthalmol 2018;11(5):898