Evidence-Based Inclusion Criteria for Cochlear Implantation in Patients With Postlingual Deafness

OBJECTIVES:This study determined the relationship between preoperative phoneme and word scores and the gain in speech perception after cochlear implantation in a large cohort of patients. The authors aimed to define evidence-based selection criteria for cochlear implantations in adults with postlingual deafness. DESIGN:This retrospective study included 364 adults with postlingual deafness who received a cochlear implant between 2000 and 2013 at the Leiden University Medical Center. The gain in speech perception observed postimplantation was compared with preoperative-aided/binaural speech perception scores, measured at 65 dB SPL in quiet. Patients who showed preoperative phoneme scores on monosyllabic words above 50% were also tested for speech perception in the presence of speech-shaped background noise, at a +5 dB signal to noise ratio. RESULTS:Speech perception in quiet improved after implantation in all except 7 patients. Average scores in quiet continued to improve, up to approximately 1 year after implantation. When participants were divided into five groups, based on preoperative speech perception scores, all groups showed gains in speech perception, except for the group with aided preoperative phoneme scores above 80%. However, all patients who were tested preoperatively with the addition of background noise showed improvements in speech perception in noise after the cochlear implantation. CONCLUSIONS:Although the decision to implant should consider individual ear differences and other factors that might apply to a particular case, based on our data, all patients with preoperative scores of either 80% (phonemes correct) or 60% (words correct) and lower in an optimal-aided situation are potential candidates for a cochlear implant, provided that their preoperative speech perception score decreases below 50% (phonemes correct) or 20% (words correct), when background noise is added at a +5 dB signal to noise ratio.