This study set out to test the null hypothesis that the distribution of blood lipid levels is the same in a population with a sensorineural hearing loss (SNHL) as in a control population. Hyperlipidaemia has been implicated as an aetiological factor in SNHL; however, the majority of reports are retrospective, lack adequate controls, or are based on a series of cases which may represent incidental findings and not a true causal relationship. In all, 1490 consecutive patients who presented to a neuro-otology clinic were studied retrospectively. This group is exceptional in that all patients had had fasting lipid profiles done regardless of their presenting problem. Those with a mean SNHL > 25 dB (0.5, 1, 2 and 4 kHz) were compared with those with hearing thresholds < or = 25 dB. An analysis of variance was also done. The study group was also compared with the National Study of Hearing data set to add external validity. Simple correlations were found between hearing thresholds and many parameters such as blood pressure, fasting glucose, triglyceride or fasting cholesterol. Analysis of variance of the neuro-otology group, controlling for variables such as age and sex, showed no significant association between hearing and blood pressure, packed cell volume, mean corpuscular volume, fasting glucose, triglyceride or fasting cholesterol. However, when general linear interactive modelling was used to analyse hearing thresholds, raised total fasting cholesterol was associated with significantly better hearing threshold levels. This study leads to a rejection of the Null hypothesis that the distribution of lipid levels is the same in a population with a hearing loss as in a control population, as hearing thresholds were found to be significantly better in those with raised cholesterol levels.
Read full abstract