Purpose: Eating disorders (EDs) are characterized by a persistent disturbance of eating patterns, leading to poor psychological and physical health. EDs' symptoms are diverse, but their biochemical manifestations can be identified in biofluids, as the gingival crevicular fluid (GCF). The development of a rapid and accurate analytical diagnostic technique, able to provide a wider comprehension of established biochemical abnormalities, would greatly assist EDs' management. Mid-infrared (MIR) spectroscopy gathers all the referred features, and is considered a fingerprint technique. In this pilot trial, the GCF discrimination of patients with EDs and controls was accessed through MIR spectroscopy, further elucidating the relevant spectral differences between both groups.
Methods: GCF was collected from 20 women with ED diagnosis and from age-matched controls. Principal component analysis and partial least squares discriminant analysis (PLSDA) were conducted on GCF MIR spectra. Different PLSDA models were considered to address the predictive capability regarding patient identification, sampling site, and presence of EDs.
Results: MIR spectroscopy was capable to discriminate GCF samples, between EDs and controls, with 84.1% of correct predictions. Regression coefficient vectors' analyses revealed that major differences were related to higher protein content in EDs.
Conclusions: Whether further studies are needed to validate the attained data, GCF MIR analysis may be regarded as an innovative, fast, and low-cost technique to assist on early diagnosis and clinical follow-up of EDs' patients.
Level of evidence: Level IV, case-control trial.
Keywords: Diagnosis; Feeding and eating disorders; Gingival crevicular fluid; Infrared; Oral; Spectrophotometry.