The outcome of cochlear implantation depends on multiple variables including the underlying health of the cochlea. Brain derived neurotrophic factor (BDNF) has been shown to support spiral ganglion neurons and to improve implant function in animal models. Whether endogenous BDNF or BDNF-regulated proteins can be used as biomarkers to predict cochlear health and implant outcome has not been investigated yet. Gene expression of BDNF and downstream signaling molecules were identified in tissue of human cochleae obtained from normal hearing patients (n = 3) during skull base surgeries. Based on the gene expression data, bioinformatic analysis was utilized to predict the regulation of proteins by BDNF. The presence of proteins corresponding to these genes was investigated in perilymph (n = 41) obtained from hearing-impaired patients (n = 38) during cochlear implantation or skull base surgery for removal of vestibular schwannoma by nanoscale liquid chromatography coupled to tandem mass spectrometry (nano LC-MS/MS). Analyzed by mass spectrometry were 41 perilymph samples despite three patients undergoing bilateral cochlear implantation. These particular BDNF regulated proteins were not detectable in any of the perilymph samples. Subsequently, targeted analysis of the perilymph proteome data with Ingenuity Pathway Analysis (IPA) identified further proteins in human perilymph that could be regulated by BDNF. These BDNF regulated proteins were correlated to the presence of residual hearing (RH) prior to implantation and to the performance data with the cochlear implant after 1 year. There was overall a decreased level of expression of BDNF-regulated proteins in profoundly hearing-impaired patients compared to patients with some RH. Phospholipid transfer protein was positively correlated to the preoperative hearing level of the patients. Our data show that combination of gene expression arrays and bioinformatic analysis can aid in the prediction of downstream signaling proteins related to the BDNF pathway. Proteomic analysis of perilymph may help to identify the presence or absence of these molecules in the diseased organ. The impact of such prediction algorithms on diagnosis and treatment needs to be established in further studies.