Assessing the volume of large cerebral arteriovenous malformations: Can the ABC/2 formula reliably predict true volume?

Background: Outcomes of stereotactic radiosurgery in the treatment of cerebral arteriovenous malformations (AVMs)are volume-dependent. The ability to estimate AVM volume has significant value in guiding AVM management. Objective: To determine whether AVM volume measurement calculated from the ABC/2 formula is accurate compared to volume calculated by a computer-assisted planimetric method for large AVMs. Methods: Retrospective review of 42 intracranial AVMs >3 cm in diameter that underwent treatment with dose-staged hypofractionated stereotactic radiotherapy (HSRT)from 2001 to 2018. Two raters independently measured pre- and post-HSRT volumes using both the ABC/2 formula and computer-assisted planimetry in a blinded fashion. Inter-rater reliability was assessed by calculation of intra-class correlation coefficient (ICC). Absolute volumes and percent volume change following HSRT as determined using the two methods were compared using paired t-tests, linear regression, and Bland-Altman plot analyses. Results: The ICC between the 2 raters for planimetric and ABC/2 volumes was 0.859 and 0.799, respectively. ABC/2 volumes, 26.1 ± 26.6 cm³, were statistically smaller than planimetric volumes, 28.6 ± 27.1 cm³ (P =.008). Despite differences, the two methods were highly correlated (R² = 0.904, linear regression). The percent volume change following HSRT was significantly greater with the ABC/2 method than compared to planimetry (P =.009). Conclusion: The ABC/2 and planimetric methods are reproducible for measuring cerebral AVM volumes. Although the ABC/2 method of volume estimation underestimates planimetric AVM volume, the high correlation between the two suggests utility of the ABC/2 method if one understands its limits, particularly with respect to estimating change in AVM volume after treatment.