Influence of Hashimoto's thyroiditis on intraoperative probe-assisted near-infrared autofluorescence detection of parathyroid glands: A cross-sectional study.

Abstract

Probe-assisted near-infrared autofluorescenc has emerged as a promising technique for intraoperative identification of parathyroid glands. However, the influence of Hashimoto's thyroiditis on detection of parathyroid glands with probe-assisted near-infrared autofluorescence remains unclear. This study aimed to evaluate the influence of Hashimoto's thyroiditis on probe-assisted near-infrared autofluorescence fluorescence signal features in type A parathyroid glands during thyroidectomy. This single-center cross-sectional study included patients who underwent thyroidectomy between June and November 2024. Clinical data, including baseline demographic and laboratory characteristics and intraoperative probe-assisted near-infrared autofluorescence measurements, were collected for analysis. Parathyroid glands were categorized into type A and type B based on their anatomic location. Probe-assisted near-infrared autofluorescence fluorescence detected values were measured intraoperatively for each suspicious parathyroid gland. Comparative analyses were performed between type A and type B parathyroid glands to evaluate differences in fluorescence detected value. Further, patients were stratified into Hashimoto's thyroiditis and non-Hashimoto's thyroiditis groups to evaluate differences in clinical characteristics. We also investigated the influence of Hashimoto's thyroiditis on the specific anatomic and fluorescence signal features of type A parathyroid glands. Univariable and multivariable linear regression analyses were conducted to identify independent variables of fluorescence-detected value and ratio (detected/baseline). To evaluate the relative contribution of covariates, variable importance was further assessed using least absolute shrinkage and selection operator regression. Lastly, we conducted receiver operating characteristic curve analysis to assess diagnostic performance of the models. Compared with the non-Hashimoto's thyroiditis group, the Hashimoto's thyroiditis group showed significantly lower probe-assisted near-infrared autofluorescence fluorescence detected values and baseline and preoperative parathyroid hormone levels (P < .05) in type A parathyroid glands. In multivariate models, thyroid peroxidase antibody and baseline were identified as independent impact variables of probe-assisted near-infrared autofluorescence fluorescence detected values (P < .001). For ratio, sex and baseline were significant impacted variables (P < .05). Least absolute shrinkage and selection operator regression revealed sex, Hashimoto's thyroiditis, thyroid peroxidase antibody, and baseline as the top predictors for detected value, whereas gender and thyroid peroxidase antibody were key for ratio. Receiver operating characteristic curve analysis indicated that the presence of Hashimoto's thyroiditis did not compromise the performance of probe-assisted near-infrared autofluorescence in identifying parathyroid glands (all-group analysis area under the curve ≥0.945). The ratio-based model demonstrated superior sensitivity and maintained high specificity, achieving 100% sensitivity in the Hashimoto's thyroiditis subgroup. Hashimoto's thyroiditis reduces probe-assisted near-infrared autofluorescence fluorescence detected value in type A parathyroid glands, yet the ratio remains stable. Despite autoimmune interference, this technique consistently enables accurate intraoperative identification of parathyroid glands, supporting its reliable use in patients with Hashimoto's thyroiditis.