Scholarlink

Purpose <p>To investigate factors correlated to retinoblastoma regression patterns defined by optical coherence tomography (OCT).</p> Methods <p>This exploratory cross-sectional study included 33 eyes of 31 cured retinoblastoma patients having 37 regressed tumors using spectral domain OCT. Three reviewers classified OCT patterns, and inter-rater reliability was assessed using Fleiss’ kappa. Associations between patterns and demographic and clinical factors were analyzed.</p> Results <p>Mean follow-up was 117.2 (± 57.3) months. Inter-rater agreement for pattern assignment was substantial (κ = 0.774). Pattern B was most common (45.9%; 95% CI, 29.4%-63.1%), followed by pattern A (29.7%; 95% CI,15.8%-46.9%) and C (24.3%; 95% CI,11.7%-41.2%). Age was significantly associated with OCT pattern distribution. In patients ≤ 10 years, pattern B predominated (70%), in those &gt; 10 years, pattern A was most frequent (47.1%), then pattern C (35.3%). Pattern C showed greater posterior tumor margin–disc distance than patterns A and B. Initial tumor height differed significantly across patterns, with pattern B showing greater height than pattern C. OCT patterns were significantly associated with clinical regression types. Type I regression predominantly showed pattern B (81.8%) and less frequently pattern A (18.2%). Type II showed equal distribution between patterns A and B. Type III corresponded exclusively to pattern B. Type IV was distributed across patterns A (36.4%), B (22.7%), and C (40.9%). All pattern C tumors were observed with type IV regression. No significant associations were observed with gender, laterality, tumor base diameter, treatment intensity, regression size, or follow-up duration.</p> Conclusion <p>Tumor anteroposterior location and initial height showed associations with OCT regression patterns.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Factors correlated to different patterns of regression of retinoblastoma defined by optical coherence tomography

  • 期刊论文

  • 发表日期：

    2026年6月1日

查看全文
  • Hisham Mohamed Khairy Mahmoud Helmy Abdel Dayem,
  • Nada Abdel Salam Abdel Aziz,
  • Othman Ali Othman Ziko,
  • Azza Mohamed Ahmed Said,
  • Noha Salah Mohamed

摘要

Purpose

To investigate factors correlated to retinoblastoma regression patterns defined by optical coherence tomography (OCT).

Methods

This exploratory cross-sectional study included 33 eyes of 31 cured retinoblastoma patients having 37 regressed tumors using spectral domain OCT. Three reviewers classified OCT patterns, and inter-rater reliability was assessed using Fleiss’ kappa. Associations between patterns and demographic and clinical factors were analyzed.

Results

Mean follow-up was 117.2 (± 57.3) months. Inter-rater agreement for pattern assignment was substantial (κ = 0.774). Pattern B was most common (45.9%; 95% CI, 29.4%-63.1%), followed by pattern A (29.7%; 95% CI,15.8%-46.9%) and C (24.3%; 95% CI,11.7%-41.2%). Age was significantly associated with OCT pattern distribution. In patients ≤ 10 years, pattern B predominated (70%), in those > 10 years, pattern A was most frequent (47.1%), then pattern C (35.3%). Pattern C showed greater posterior tumor margin–disc distance than patterns A and B. Initial tumor height differed significantly across patterns, with pattern B showing greater height than pattern C. OCT patterns were significantly associated with clinical regression types. Type I regression predominantly showed pattern B (81.8%) and less frequently pattern A (18.2%). Type II showed equal distribution between patterns A and B. Type III corresponded exclusively to pattern B. Type IV was distributed across patterns A (36.4%), B (22.7%), and C (40.9%). All pattern C tumors were observed with type IV regression. No significant associations were observed with gender, laterality, tumor base diameter, treatment intensity, regression size, or follow-up duration.

Conclusion

Tumor anteroposterior location and initial height showed associations with OCT regression patterns.