ABSTRACT
BACKGROUND: Multiple differentials exist for pediatric liver tumors under 2 years. Accurate imaging diagnosis may obviate the need for tissue sampling in most cases. OBJECTIVE: To evaluate the imaging features and diagnostic accuracy of computed tomography (CT) in liver tumors in children under 2 years. METHODS: Eighty-eight children under 2 years with treatment naive liver neoplasms and baseline contrast-enhanced CT were included in this institutional review board approved retrospective study. Two blinded onco-radiologists assessed these tumors in consensus. Findings assessed included enhancement pattern, lobulated appearance, cystic change, calcifications, central scar-like appearance, and metastases. The radiologists classified the lesion as hepatoblastoma, infantile hemangioma, mesenchymal hamartoma, rhabdoid tumor, or indeterminate, first based purely on imaging and then after alpha-fetoprotein (AFP) correlation. Multivariate analysis and methods of comparing means and frequencies were used for statistical analysis wherever applicable. Diagnostic accuracy, sensitivity, and positive predictive values were analyzed. RESULTS: The mean age of the sample was 11.4 months (95% CI, 10.9-11.8) with 50/88 (57%) boys. The study included 72 hepatoblastomas, 6 hemangiomas, 4 mesenchymal hamartomas, and 6 rhabdoid tumors. Presence of calcifications, multilobular pattern of arterial enhancement, lobulated morphology, and central scar-like appearance was significantly associated with hepatoblastomas (P-value < 0.05). Fourteen out of eighty-eight lesions were called indeterminate based on imaging alone; six lesions remained indeterminate after AFP correlation. Pure radiology-based diagnostic accuracy was 81.8% (95% CI, 72.2-89.2%), which increased to 92.1% (95% CI, 84.3-96.7%) (P-value > 0.05) after AFP correlation, with one hepatoblastoma misdiagnosed as a rhabdoid tumor. If indeterminate lesions were excluded for biopsy, the accuracy would be 98.8% (95% CI, 93.4-99.9%). CONCLUSION: CT had high accuracy for diagnosing liver neoplasms in the under 2-year age population after AFP correlation. Certain imaging features were significantly associated with the diagnosis of hepatoblastoma. A policy of biopsying only indeterminate lesions after CT and AFP correlation would avoid sampling in the majority of patients.
ABSTRACT
AIM: Complimentary use of Liver directed therapies (LDTs) with systemic chemotherapy has improved oncologic outcomes in colorectal liver metastasis (CRLM). We analysed institutional results of multimodality management. Methods: Retrospective analysis of prospectively maintained database of CRLM patients managed with LDT including surgical resection, Ablation, Transarterial chemoembolization (TACE) or Transarterial radioembolization (TARE) between November 2011 to March 2020. Management plan was decided in multidisciplinary meeting. Resectable tumours underwent surgical resection or ablation or both in some cases. Borderline resectable or unresectable disease was treated with down staging chemotherapy or TACE/TARE followed by resection or ablation. All patients received adjuvant chemotherapy. Factors influencing survival were analysed. Results: Out of total 375 patients, surgery alone was done in 191 (50.93%) patients while surgery with other LDT in 26 patients (6.93%). Ablation alone was done in 100 (26.66%) whereas TACE/TARE were done as standalone treatment in 21 (5.6%) and 7 (1.86%) patients respectively. TACE + ablation was done in 28 (7.46%) and TARE + ablation was done in 2(0.53%) patients.5-year Overall Survival(OS) was 49.8% while Event free survival(EFS) was 21.4%. The median OS and EFS for surgical group was significantly better than non-surgical group (78 V/s 39 months; p<0.05 and 20 V/s 15 months p <0.005). The resectable (78 months) group had better median OS as compared to borderline resectable and Unresectable group (39 months and 29 months). Male gender, resectable disease and surgical intervention were associated with improved OS. Conclusion: Although surgery remains the mainstay of treatment, complementary use of non-surgical LDT with systemic therapy offers possibility of good outcomes in advanced liver limited disease. Our experience highlights the impact of multidisciplinary care in optimizing CRLM treatment.
ABSTRACT
Fat necrosis of the breast is a benign non-suppurative inflammation of the adipose tissue and often mimics breast cancers, posing a diagnostic challenge for the clinician and radiologist. It has a myriad of appearances on different imaging techniques, ranging from the pathognomic oil cyst and benign dystrophic calcifications to indeterminate focal asymmetries, architectural distortions, and masses. A combination of different modalities can assist a radiologist in reaching a logical conclusion to avoid unnecessary interventions. The aim of this review article was to provide a comprehensive literature on the various imaging appearances of fat necrosis in the breast. Although a purely benign entity, the imaging appearances on mammography, contrast-enhanced mammography, ultrasound, and magnetic resonance imaging can be quite misleading, especially in post-therapy breasts. The purpose is to provide a comprehensive and all-inclusive review on fat necrosis with a proposed algorithm allowing a systematic approach to diagnosis.
