Anthropometry- is it useful in assessment for obstructive sleep apnoea (OSA)?

Cardinal features of OSA are snoring, witnessed apnoeas and diurnal somnolence. Clinical signs are absent from this list and apart from measures of obesity, their use in evaluating OSA has not been systematically examined. Aim: To examine the usefulness of simple anthropométrie indices in assessing patients with suspected OSA. Methods: 480 consecutive newly referred patients with suspected OSA requiring polysomnography were approached of whom 471 (363 male, 108 female) participated. Height; weight; neck (nc), hip (he) and waist (we) circumferences; mandibular ramus and base lengths; degree of neck extension from head neutral to maximum; cricomental distance; overjet (George gauge); and palatal appearance (Mallampati score) were measured in each patient. The influence of these parameters on OSA was determined using multiple linear regression analysis with respiratory disturbance index (RDI) as the dependent variable. Results: There was covariance between weight, body mass index (bmi), we and he in males as well as nc in females. When weight, we and he and, in females, nc were eliminated, the statistical model remained significantly predictive of OSA for both males (r² =0.30,p<0.001) and females (r²=0.42,p<0.001).The most important individual predictors were bmi (r²=0.22,p<0.001) and nc for males (r²=0.22,p<0.001) and bmi (r²=0.26,p<0.001) and limited neck extension in females (r²=0. 1 2,p<0.001). Conclusions: 1) Anthropometry alone explains 30% and 42% of variability of RDI in males and females respectively. 2) Indices of truncal obesity show significant covariance except for nc in males which appears to have predictive power that is independent of degree of obesity. 3) Decreased neck extension appears a useful anthropométrie predictor of RDI in females.