To date, there is not a validated version of Karolinska Sleepiness Scale for Mexican population. Objective: To study the validity of convergence and diagnosis of a Spanish version of the EKS-M. Methods: It was a prospective psychometric study in which the EKS-M was applied to a sample of 87 patients (54% of the male sex) with a polysomnographic diagnosis of sleep-related breathing disorders, insomnia or periodic limb movement disorders during sleep. The study was done at the Sleep Disorders Clinic of the Universidad Nacional Autónoma de México. The EKS-M was applied three times: before the PSG record, before the placement of the CPAP and at the conclusion of the PSG record. Pearson correlation test was used to determine the relationship between EKS-M and PSG features; and a value of p < 0.5 was used as the cut-off point of the statistical significance. A positive relationship was found between the EKS-M prior to the PSG with the Epworth Sleepiness Scale (r = 0.307, p <0.10), the average duration of central apnea (r = 0.234, p <0.047) and number of snoring (r = 0.220, p <0.33), and a negative correlation with latency at the onset of sleep (r = -0.223, p <0.036). Regarding the post-PSG EKS-M we found a positive relationship slow wave sleep percentage before CPAP tritation (r = 0.451, p <0.023), number of hypopneas (r = 0.339, p <0.020), number of snoring (r = 0.347, p <0.005) and the index of periodic limb movement (r = 0.388, p <0.025). The EKS-M was a helpful diagnostic tool for sleepiness severity, with very acceptable validity indicators (concerning to more sleepiness severity and PSG features such as shortening sleep onset latency, plus it is a method of rapid and economical application. This will allow to obtain large samples in a short time. This research had no support.