The ability of magnetic resonance imaging (MRI) to detect very small acoustic tumors has triggered many to rethink the use of auditory brainstem response (ABR) in the screening of acoustic tumors. To assess ABR accuracy, we conducted a retrospective study of 388 surgically treated patients. Of these patients, 111 had complete databases including both preoperative MRIs and ABRs. The ABR was abnormal by wave V interaural latency difference in 106 (95%) of the cases. Although our overall sensitivity was 95%, sensitivity varied according to tumor size. ABR was abnormal or absent for all tumors (100%) larger than 2 cm in diameter, for 98% of tumors 1.1 to 2 cm in diameter, and for only 89% of tumors less than or equal to 1 cm in diameter. Ramifications of this in the decision-making process are presented. Criteria for cut-off values are also discussed.