Although the relationship between hearing loss and hyperlipoproteinemia (HLP) or diabetes mellitus (DM) has been shown in many clinical investigations, this concept is still controversial. A prospective study was designed to search for the existence of subclinical auditory dysfunction related to HLP and DM by transient-evoked (TEOAEs) and distortion-product otoacoustic emissions (DPOAEs) in patients with hearing levels better than 30 dB. Evoked otoacoustic emissions were utilized to investigate subclinical auditory dysfunction. Fifteen hypercholesterolemic patients (28 ears), 21 hypertriglyceridemic patients (42 ears) and 21 DM patients (40 ears) were eligible for investigation. The results of the DPOAEs and TEOAEs of the study groups were compared with the control group composed of individuals with similar ages and with normal blood lipids and glycemia. This group consisted of 22 people (44 ears). There was no difference in the existence of TEOAEs at all frequencies among the groups (P>0.05). No differences were found in the amplitudes of the DPOAES between the groups except at 4 kHz (P>0.05). The difference was caused by the hypertriglyceridemia group (P=0.014) and the non-insulin-dependent diabetes mellitus (NIDDM) group (P=0.012) when compared with the control group. The mean DPOAE amplitudes of the hypertriglyceridemic and NIDDM groups at 4 kHz were higher, than those of the control group. The decreased DPOAE amplitudes at 4 kHz in hypertriglyceridemic and diabetic patients without clinical findings are compatible with the sensorineural hearing loss observed with hyperviscosity and increased noise susceptibility, as was shown before in these patients. Longitudinal investigations should be performed with otoacoustic emissions to help with the early prediction of the prospective effects of HLP and DM on the auditory system.