Xiao Ma, Qi-Jun Xia, Yun-Tao Song, Tian-Xiao Wang
BAA thyroidectomy offers superior cosmetic satisfaction, less pain, and reduced blood loss compared with COA, with benefits persisting after rigorous confounding adjustment. The preoperative model demonstrated clinical utility in personalizing patient selection for BAA. BAA is a safe and highly satisfactory option for selected patients with papillary TC. However, patient selection based on cosmetic preference introduces unmeasured confounding by preoperative cosmetic expectations, which may influence subjective outcomes.
PURPOSE: To compare the clinical outcomes of bilateral areolar endoscopic approach (BAA) vs. conventional open approach (COA) thyroidectomy for papillary thyroid carcinoma (TC) and to evaluate the added value of propensity score matching (PSM) and decision curve analysis (DCA).
METHODS: Eighty-eight female patients were retrospectively divided into BAA (n = 42) and COA (n = 46) groups according to aesthetic demand. Perioperative outcomes, pain scores, and cosmetic satisfaction (visual analogue scale, VAS) were compared. Propensity score full matching (full + glm) was performed to adjust for baseline confounders. A preoperative prediction model for high cosmetic satisfaction (VAS>9) after BAA was developed and assessed by DCA.
RESULTS: Patients in the BAA group were significantly younger (29.0 ± 5.7 vs. 43.5 ± 8.6 years, P = 0.001). Compared with COA, BAA was associated with less intraoperative blood loss (15.3 ± 8.4 vs. 30.2 ± 11.9 mL, P < 0.001), lower postoperative pain (4.13 ± 2.54 vs. 5.15 ± 2.46, P < 0.001), and markedly higher cosmetic satisfaction (9.52 ± 1.18 vs. 7.82 ± 1.76, P < 0.001), despite longer operative time (97.6 ± 13.9 vs. 76.4 ± 12.2 min, P < 0.001). No significant differences were observed in lymph node yield, hospital stay, or complications. After propensity score full matching that substantially reduced confounding (especially age), BAA retained a 1.3-point higher cosmetic satisfaction score (95% CI 0.98-1.64, P = 7.93 × 10⁻1⁵) and 14.9 mL less blood loss (P = 1.22 × 10⁻⁹). DCA in the BAA cohort showed that the prediction model incorporating age, BMI, tumor size, and unilateral procedure provided positive net benefit, identifying 30-35 additional truly highly satisfied patients per 1,000 across plausible risk thresholds.
CONCLUSION: BAA thyroidectomy offers superior cosmetic satisfaction, less pain, and reduced blood loss compared with COA, with benefits persisting after rigorous confounding adjustment. The preoperative model demonstrated clinical utility in personalizing patient selection for BAA. BAA is a safe and highly satisfactory option for selected patients with papillary TC. However, patient selection based on cosmetic preference introduces unmeasured confounding by preoperative cosmetic expectations, which may influence subjective outcomes.