gms | German Medical Science

Objective: The assessment of tumor size is essential in the management of patients suffering from vestibular schwannomas (VS). The aim of our study was to determine the effect of the following items on magnetic resonance (MR) measurements of these tumors: a) the inter-rater reliability (IRR), b) the intra-rater variability, c) the orientation of acquired MR data (axial versus coronal), and d) the correlation of one-dimensional and volumetric measurements.

Method: We selected 20 patients who had both axial and coronal post-contrast T1-weighted MRI data sets available with the same slice thickness in both orientations. All patients suffered from either sporadic (n=8) or NF2-associated (n=12) VS. Five of the NF2-associated VS had previously been operated on, one had undergone radiosurgery, one had undergone surgery and radiation and another one had undergone surgery and bevacizumab treatment. Tumor volumes were determined by volumetric extrapolation after segmentation (iPlan software, BrainLAB, Munich). Diameters were measured as largest diameter across the whole tumor including the inner auditory canal (GtD). In order to find a term to calculate tumor volume from GtD we fitted a series of functions based on linear and polynomial regression. All measurements were performed by two independent investigators. After an interval of 6 months investigator 1 performed all measurements a second time in order to determine intra-rater variability.

Results: The IRR was determined based on the intra-class correlation coefficient, which was 0.998 for volumetric measurements and 0.950 for GtD. The smallest detectable growth difference (SDD%) between both raters was 21.22% for volumetric and 21.19% for linear measurements. Regarding the intra-rater variability we found a SDD% of 17.48% for volumetric measurements and 16.72% for GtD. There was a strong correlation between measurements on axial and those on coronal data sets (Pearson's ρ=0.999 for both investigators). The SDD% was 19.02% for rater 1 and 19.17% for rater 2. The best fitting function with the highest coefficient of determination showed r2=0.79.

Conclusions: Volumetric tumor size assessment is superior to one-dimensional measurement. For longitudinal follow-up, it is safe to obtain tumor volumes from data sets of varying orientation. A strict MRI protocol for follow-up investigations should be adhered to in order to minimize measuring errors.