Premium
Anthropometric Measurements as a Screening Test for Carpal Tunnel Syndrome: Receiver Operating Characteristic Curves and Accuracy
Author(s) -
Mondelli Mauro,
Curti Stefania,
Farioli Andrea,
Aretini Alessandro,
Ginanneschi Federica,
Greco Giuseppe,
Mattioli Stefano
Publication year - 2015
Publication title -
arthritis care and research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.032
H-Index - 163
eISSN - 2151-4658
pISSN - 2151-464X
DOI - 10.1002/acr.22465
Subject(s) - medicine , receiver operating characteristic , carpal tunnel syndrome , anthropometry , waist , wrist , cutoff , body mass index , logistic regression , circumference , confidence interval , likelihood ratios in diagnostic testing , area under the curve , surgery , mathematics , physics , geometry , quantum mechanics
Objective To identify optimal cutoff values for body, hand, and wrist measurements in order to correctly identify individuals with carpal tunnel syndrome (CTS), using receiver operating characteristic (ROC) curves. Methods We enrolled patients with CTS and control subjects at a 1:2 ratio, regardless of age and sex. The diagnosis of CTS was based on clinical findings and delayed distal conduction velocity of the median nerve. The anthropometric measurements included weight, height, waist circumference, hip circumferences, wrist depth/width, third digit length, and palm length/width. Obesity indicators and hand/wrist ratios were calculated. Area under the ROC curve (AUC), sensitivity, specificity, and likelihood ratios were calculated separately according to sex. To assess the role of multiple anthropometric measurements, we fit multivariable logistic regression models including age, wrist ratio, shape index, body mass index, and waist‐to‐hip ratio. Results The study group comprised 1,117 subjects (250 female patients and 474 female controls; 120 male patients and 273 male controls). In women, the accuracy of all anthropometric measures was low (AUC ≤0.64). In men, the accuracy of the hand ratio, shape index, and wrist‐to‐palm ratio was moderate (AUC = 0.75). The estimates from the multivariable models confirmed the well‐known associations between the selected variables and the risk of CTS, but the use of multiple predictors did not dramatically improve the diagnostic performance observed for single anthropometric indexes. Conclusion In clinical practice, the cutoff values for many anthropometric measurements have limited value as tools for the diagnosis of CTS.