AIM: Direct neoplastic invasion of esophageal inlet is an uncommon but significant sequela of advanced head and neck carcinomas. The aim of this study was to seek an optimal CT or MRI criterion for determining the neoplastic esophageal inlet involvement in order to help tumor staging and surgical planning. METHODS: CT and MRI of 78 head and neck tumor cases were investigated retrospectively. At the level of the esophageal inlet on axial CT and MRI scans, the distance between the posterior aspect of cricoid cartilage and the anterior aspect of vertebra (d-CV) was measured by two senior radiologists who were unaware of clinical findings. Then, according to pathologie evidence and follow-up findings, these patients were divided into patients group, including 32 cases with neoplastic invasion of esophageal inlet and control group, including 46 cases without neoplastic esophageal inlet involvement. The statistical difference based on d-CV between the two groups was determined. The optimal criterion of d-CV on CT or MRI was assessed and lts accuracy, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were evaluated respectively. RESULTS: In control group, d-CV at the esophageal inlet level was 0.94±0.15 cm on axial CT and 0.91±0.18 cm on axial MRI, whereas in patient group, d-CV was 1.24±0.32 cm on CT and 1.31±0.36 cm on MRI. There was a statistical significance in d-CV between the two groups on CT and MRI modalities (P<0.01). d-CV greater than 1.0 cm was the typicall feature of neoplastic invasion of the esophageal inlet with 73% sensitivity, 83% specificity, 79% accuracy, 76% PPV, 80% NPV on CT and 84% sensitivity, 77% specificity, 80% accuracy, 70% PPV, 88% NPV on MRI respectively. CONCLUSION: Except for other CT and MR imaging features of neoplastic invasion of esophageal inlet, d-CV greater than 1.0 cm is an optimal adjunct criterion for esophageal inlet invasion by advanced head and neck carcinomas. AIM: Direct neoplastic invasion of esophageal inlet is an uncommon but significant sequela of advanced head and neck carcinomas. The aim of this study was to seek an optimal CT or MRI criterion for determining the neoplastic esophageal inlet involvement in order to help tumor staging and surgical METHODS. CT and MRI of 78 head and neck tumor cases were investigated retrospectively. At the level of the esophageal inlet on axial CT and MRI scans, the distance between the posterior aspect of cricoid cartilage and the anterior aspect of vertebra (d- CV, was measured by two senior radiologists who were unaware of clinical findings. Then, according to pathologie evidence and follow-up findings, these patients were divided into patients group, including 32 cases with neoplastic invasion of esophageal inlet and control group, including 46 cases without neoplastic esophageal inlet involvement. The statistical difference based on d-CV between the two groups was determined. The optimal c riterion of d-CV on CT or MRI was assessed and lts accuracy, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were evaluated respectively. RESULTS: In control group, d- CV at the esophageal inlet level was 0.94 ± 0.15 cm on axial CT and 0.91 ± 0.18 cm on axial MRI, while in patient group, d-CV was 1.24 ± 0.32 cm on CT and 1.31 ± 0.36 cm on MRI. There was a statistical significance in d-CV between The two groups on CT and MRI modalities (P <0.01). d-CV greater than 1.0 cm was the typicall feature of neoplastic invasion of the esophageal inlet with 73% sensitivity, 83% specificity, 79% accuracy, 76% PPV, 80 % NPV on CT and 84% sensitivity, 77% specificity, 80% accuracy, 70% PPV, 88% NPV on MRI respectively. CONCLUSION: Except for other CT and MR imaging features of neoplastic invasion of esophageal inlet, d-CV greater than 1.0 cm is an optimal adjunct criterion for esophageal inlet invasion by advanced head and neck carcinomas.