Thyroid evaluation is frequently performed in patients with sleep apnea because of a suspected causal relationship between hypothyroidism and obstructive sleep apnea (OSA). The aim of this study was to determine the actual prevalence of hypothyroidism in patients referred for polysomnography and evaluate whether its rate was higher in patients with OSA than those without OSA. Ultrasensitive thyroid stimulating hormone (TSH) was performed on 255 of 279 consecutive patients referred for polysomnography from the neurology service of a large HMO. Hypothyroidism was detected in 1.6% (4/243) of all patients, 1.5% (3/194) of patients referred to evaluate OSA, and 2.0% of patients referred to evaluate the presence of periodic leg movement disorder (PLMD)/narcolepsy/parasomnia. There was no significant difference in rates of hypothyroidism in patients with documented OSA (2.9%, 3/103) compared to those without OSA (0.7%, 1/135). Two of the four patients with elevated TSHs had previously documented hypothyroidism and were on thyroxine replacement. Rates of hyperthyroidism were as high or higher than those of hypothyroidism in all groups. We conclude that thyroid screening does not appear to be appropriate for patients with suspected, or confirmed, OSA in the absence of signs or symptoms consistent with hypothyroidism or unless they are in a high risk group (women over the age of 60).