| Objective To investigate the difference and consistency of multimodality magnetic resonance imaging(MRI)in measuring gross tumor volume(GTV)for nasopharyngeal carcinoma(NPC).Methods Between December 2012 and March 2013,45 patients with histopathologically confirmed NPC were recruited.All patients underwent conventional,dynamic contrast-enhanced MRI and diffusion-weighted imaging(DWI).Two radiologists independently outlined tumor profile and blinded to each other.Tumor volumes measured on axial T1WI,axial T2WI,apparent diffusion cofficient(ADC)map were calculated with standard techniques of tumour area multiplied by the layer thickness and spacing.The retropharyngeal lymph nodes were included in GTV when no clear boundary between nasopharyngeal primary tumor and retropharyngeal nodes.The intraclass correlation coefficient(ICC)was used to evaluate the inter-observer agreement.One-way analysis of variance was used to evaluate the differences among three groups.Bland-Altman assessment was used to evaluate the consistency of multimodality MRI in measuring GTV.Results The GTV of T1+c,T2WI and ADC group were(21.2±14.1),(20.7±14.2),(24.0±16.5)cm3,respectively.Tl+c volumes were not significantly different from T2WI volumes or from ADC map volumes(P>0.05).Inter-observer agreement was found good consistency between the two radiologists for each ICC was greater than 0.81.ICC of inter-observer agreement on GTV was 0.87 of Tl+c,0.89 of T2WI,0.83 of ADC,respectively.The mean differences,average value and 95%limits of agreement between the group of T2WI and T1+C,ADC and T2WI,ADC and T1+c were 2.05,3.04,3.62 cm3,and 20.1,22.6,22.4 cm3,and(-5.5,6.4),(-1.9,7.5),(-5.6,12.1)cm3,respectively.Conclusion The multimodality MRI have poor consistency in measuring GTV for NPC.They cannot replace each other.Enhanced T1WI combined with T2WI and DWI may determine a relatively accurate range of target volume for NPC. |
PDF Full Text Request