[Validation of the Spanish version of the Epworth Sleepiness Scale in patients with a sleep apnea syndrome]

Abstract

A Spanish version of the Epworth Sleepiness Scale (ESS-Sp) was developed by translation, back-translation, formal discussion, and a meeting of researchers with a group of patients with sleep apnea syndrome (SAS). The translated questionnaire was then tested in 345 patients, 275 with SAS at various levels of severity and 70 without SAS. Significant differences existed between the two groups as to age (53 +/- 11 years versus 47 +/- 13, p < 0.001) and BMI (32 +/- 5 versus 29.5 +/- 5, p < 0.001). Patients with SAS had significantly higher scores (14 +/- 5) than did those without SAS (10 +/- 5) (p < 0.001). Reproducibility was tested in 146 patients (113 SAS and 33 non-SAS), with no significant differences found among patients with SAS (14.9 +/- 5 versus 14.2 +/- 5, p = n.s.); significant differences in BMI were found, however, among the 33 non-SAS patients (12 +/- 5 versus 10 +/- 5, p < 0.01). Total scores and individual item scores were related in both groups. Likewise, each item was related to total score in patients with SAS. Sensitivity to post-treatment changes was assessed in 77 SAS patients, with initial scores of 16 +/- 4 seen to decrease to 4 +/- 3 after continuous positive airway pressure. ESS-Sp scores over 10 were recorded for 85% of patients with SAS: 78% of those with mild SAS, 85% of those with moderate disease and 92% of those whose SAS was severe. Significant inter-group differences were found upon applying a test of variance (p < 0.001). Differences continued to be detected when multiple correlations were looked for, with differences increasing with severity. SAS patients with ESS-Sp level one scores (< 10) had lower apnea-hypopnea indices (AHI) (35 +/- 18 versus 42 +/- 20, p < 0.05), lower desaturation levels (21 +/- 21 versus 34 +/- 28, p < 0.01) and higher minimum saturation (80 +/- 10 versus 75 +/- 12, p < 0.05), with no differences in age or BMI. A significant correlation was found between ESS-Sp score and respiratory variables recorded during polysomnography: AHI, r = 0.23 (p < 0.001); percent time in apnea-hypopnea, r = 0.18 (p < 0.01); desaturation index, r = 0.27 (p < 0.01) and minimum saturation (r = -0.14, p < 0.05). We conclude that the Spanish version of the ESS is equivalent to the original, is reproducible in patients with SAS, sensitive to post-treatment changes and seems to discriminate level of severity, showing correlation with polysomnograph variables.