Concerns about the Validation of the Berlin Questionnaire and American Society of Anesthesiologist Checklist as Screening Tools for Obstructive Sleep Apnea in Surgical Patients

In the May 2008 issue of Anesthesiology, Chung et al . published an article where they describe the validation of the Berlin questionnaire and the American Society of Anesthesiologist checklist as screening tools for obstructive sleep apnea in surgical patients, and compare them with the STOP questionnaire.1The authors conclude that both the STOP and American Society of Anesthesiologists checklist were able to identify patients who were likely to develop postoperative complications. However, after taking a look at the results, I believe that it would have been more accurate to mention respiratory complications in particular. Moreover, when reviewing the odds ratios for effectors on the incidence of postoperative complications, I am concerned to find that the confidence intervals for both the STOP and American Society of Anesthesiologists checklists include the null value.2In contrast, the odds ratio for the STOP-Bang questionnaire presents a confidence interval that does not comprise the null value.With reference to the potential limitations for this study described by the authors, I agree with them about the possible bias associated with self-selection of patients. Only 416 (17%) of 2,467 patients gave consent to participate in a polysomnographic study, whereas finally 211 (8.6% of the total population) showed up to undergo it. Another issue is, when reading the analysis of those 211 patients, there is little valuable information left about their preexisting conditions, such as number of smokers, type of surgery and anesthesia technique given,3or patients suffering from asthma or other pulmonary diseases, which could have been desirable to discuss when comparing the higher incidence of respiratory complications among patients with higher scores in the questionnaires. Knowing more about preexisting morbidities might have allowed classifying patients to make comparisons between them in further multivariate analyses.At the same time, following the requirement in one hospital to closely monitor patients with an apnea-hypopnea index greater than 30, the authors did not find this variable to be a risk factor for postoperative complications. I would want to know what result would have been obtained had those patients been excluded from the analysis.To sum up, I believe that this study presents some unsatisfactory points that hamper the conclusions given and deserve to be addressed in more detail.*Clinica Universitaria, University of Navarra, Spain. jrpvaldi@unav.es