Premium
Diagnosis of Postlaminar Optic Nerve Invasion in Retinoblastoma With MRI Features
Author(s) -
Li Zhenzhen,
Guo Jian,
Xu Xiaolin,
Wang Yongzhe,
Mukherji Suresh Kumar,
Xian Junfang
Publication year - 2020
Publication title -
journal of magnetic resonance imaging
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.563
H-Index - 160
eISSN - 1522-2586
pISSN - 1053-1807
DOI - 10.1002/jmri.26961
Subject(s) - medicine , receiver operating characteristic , confidence interval , coronal plane , odds ratio , optic nerve , magnetic resonance imaging , sagittal plane , retinoblastoma , diagnostic odds ratio , univariate analysis , area under the curve , nuclear medicine , population , radiology , pathology , multivariate analysis , ophthalmology , biochemistry , chemistry , environmental health , gene
Background Retinoblastomas (RBs) with postlaminar optic nerve invasion (PLONI) increases the risk of local recurrence or systemic metastasis. Most MRI studies on PLONI focused on optic nerve enhancement and tumor size, with reported relatively high specificity (84–98%) but low sensitivity (37–78%), which shows room for improvement, especially with regard to sensitivity. Purpose To evaluate the diagnostic performance for detecting PLONI with RB using MRI features. Study Type Retrospective. Population Fifty patients with histopathologic PLONI and 70 patients without PLONI of RB. Field Strength/Sequence 1.5T and 3.0T, precontrast axial T 1 ‐weighted and T 2 ‐weighted, postcontrast axial, coronal and oblique‐sagittal T 1 ‐weighted. Assessment The eyes were histopathologically analyzed and the preoperative MRI features of the eyes were independently evaluated by three observers. Statistical Tests MRI features suggesting the presence of PLONI were identified by univariate and multivariable analysis. Receiver operating characteristic curve (ROC) and the area under the curve (AUC) were used to analyze diagnostic performance. Results Significant independent diagnostic factors for PLONI include: 1: Bilateral tumor (odds ratio [OR], 15.32; 95% confidence interval [CI]: 1.63–143.51); 2: Tumor with total coverage of the optic disk (OR, 6.43; 95% CI: 1.04–39.79); and 3: Optic nerve enhancement (OR, 8.43; 95% CI: 3.50–20.31). On the other hand, isointense signal of tumor on T 2 WI (OR, 0.30; 95% CI: 0.12–0.75) was an independent diagnostic factor in excluding PLONI. ROC analysis showed AUC of 0.84 (95% CI: 0.77–0.91, P < 0.0001) for PLONI. Based on the cutoff of maximum Youden index, the sensitivity, specificity, and accuracy were 82%, 73%, and 77%, respectively. Data Conclusion MRI features of RB showed a strong association with PLONI. The model of MRI features demonstrated promising diagnostic performance in detecting PLONI. Level of Evidence: 4 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2020;51:1045–1052.