Skip to main content

Table 4 Prognostic factors of visual acuity deterioration expressed in logMar from baseline value using linear regression on the 61 patients with available data on visual outcomes

From: Oncologic and visual outcomes after postoperative proton therapy of localized conjunctival melanomas

 

Bivariate analysesa

Multivariate analysis a ∆

Estimationâ—Š

p-value

Estimationâ—Š

p-value

female (vs. male)

0.046(0.164)

0.781

  

Age (years)

0.010(0.005)

0.072

0.01 (0.004)

0.008

Precancerous lesion nevus vs. de novo PAM vs. de novo

0.146(0.180)

0. 365 (0.184)

0.423

0.052

0.243 (0.168)

0.285 (0.126)

0.153

0.028

Clinical stage

 T2 vs. T1

0.040(0.214)

0.853

  

 T3 vs. T1

0.524 (0.333)

0.121

  

Pathological stage

 pT2 vs. pT1

0.031(0.203)

0.880

  

 pT3 vs. pT1

0.723(0.375)

0.059

  

Multifocal

0.768(0.202)

< 0.001

0.554 (0.157)

< 0.001

Quandrangular involvement from 90 to 180° vs < 90°

> 180° vs < 90°

1.446(0.284) 0.246(0.146)

0.098 < .001

0.176 (0.115) 0.954 (0.237)

0.130 < 0.001

Epithelioid type

0.123(0.166)

0.461

  

Margins

 R1 vs.R0

0.041(0.181)

0.8193

  

 R2 vs.R0

−0.302(0.251)

0.2329

  

Number of mitoses < 2

−0.003(0.159)

0.986

  

Ulceration

0.197(0.236)

0.407

  

Thickness < 2 mm

0.147(0.248)

0.556

  

Diameter < 7 mm

−.020(0.161)

0.899

  

Cryotherapy

0.585(0.149)

< 0.001

0.461 (0.11)

< 0.001

Mitomycin

−0.302(0.170)

0.082

  

Follow-up time (years)

0.007(0.013)

0.607

  
  1. aAdjusted on visual acuity at baseline
  2. ∆ associated R-square (percentage of explained variance) was 0.64
  3. â—ŠEstimation expressed as estimated beta value and standard deviation from linear regression. For a qualitative parameter, a beta value of 0.3 logMAR was considered as significant deterioration in visual acuity
  4. Abbreviations: T: tumor, R0: complete resection, R1: microscopic resection, R2: macroscopic resection, HR: hazard ratio; CI: Confidence interval; vs.: versus, primary acquired melanosis (PAM)