Search

Objective To evaluate the utility of magnetic resonance images (MRI) in the selection of treatment procedures for intermediate–advanced laryngeal cancers. Material and method This study included patients with histologically proven laryngeal squamous cell carcinoma defined as cT3 and T4a at our tertiary academic care hospital. All scans were evaluated by two radiologists experienced in head and neck cross-sectional studies. Signal patterns in MRI sections of laryngeal compartment subsites were delineated as T1w, T2w hyperintensity, and T2w intermediate signals, and were compared with the postoperative pathological results. Sensitivity, specificity, positive predictive values (PPV), and negative predictive values (NPV) were calculated. Results The study included 51 patients with a mean age of 62.55 ± 9.14 (range, 45–80) years. Tumor was glottic in 12 (23.5%) patients, supraglottic in 19 (31.4%) patients, glottic-supraglottic in 11 (21.6%) patients, transglottic in 9 (27%) patients. The posterior paraglottic space had the strongest specificity of MRI according to tumor infiltration in the histologic analysis (specificity 96.9% and sensitivity 78.6%). The specificity of MRI was poor for tumor infiltration in thyroid cartilage (specificity 70.0%). Spearman's test demonstrated that there was a statistically significant correlation between the MRI-based prediction scores of all subunites and the findings of histopathologic analyses (mean±SD: 4.96±4.46–5.53±4.38, respectively, R2: 0.711, p Conclusion The high specificity values of the predictions, which were MRI-based in all subsites, indicated that MRI could provide an important contribution for defining tumor infiltration and the presurgical assessment of patients with tumors of the larynx.

Aims and background Sentinel lymph node biopsy has been accepted as a standard procedure for early stage breast cancer. In this retrospective analysis, the results obtained with different methodological approaches using radiocolloid with or without blue dye were examined. Methods A total of 158 sentinel lymph node biopsies were performed in 152 patients. Group A (85 patients) underwent lymphatic mapping using a combination of periareolar intradermal radiocolloid and subareolar blue dye injections. Group B (73 patients) underwent only periareolar intradermal radiocolloid injection. One large tin colloid and two small radiocolloids (nanocolloid of serum albumin -NC- and colloidal rhenium sulphide -CS-) were used. Results Successful lymphatic mapping was attained in 157 of 158 procedures (99.4%). Radiocolloids localized sentinel lymph nodes in 99.4% and blue dye in 75.3% of the cases. The number of sentinel lymph nodes removed was greater in nanocolloid and colloidal rhenium sulphide groups ( P ≤0.05). Among 60 metastatic sentinel lymph nodes, frozen section analysis using hematoxylin and eosin staining failed to detect 1 macro- and 10 micrometastasis. Radiocolloid uptake was higher in sentinel lymph nodes accumulating blue dye (1643 ± 3216 counts/10 sec vs 526 ± 1284 counts/10 sec, P Conclusions Radiocolloids were more successful than blue dye in sentinel lymph node detection. More sentinel lymph nodes were harvested with small colloids, but different sized radiocolloids were similarly successful. Sentinel lymph nodes having higher radiocolloid uptake tended to accumulate blue dye more frequently. Sentinel lymph nodes manifested higher count rates when a larger colloid was used. Frozen section was very successful in detecting macrometastatic disease in sentinel lymph nodes, but the technique failed in most of the micrometastates.