Overview
This single-center, prospective, three-phase observational cohort quantifies the relationship between indocyanine green (ICG) fluorescence and early parathyroid function. Adult patients scheduled for unilateral thyroid lobectomy with autotransplantation of the ipsilateral inferior parathyroid gland are enrolled. After intravenous ICG 25 µg/kg, peak fluorescence of the superior parathyroid gland and the common carotid artery (CCA) is recorded at 60-120 seconds to calculate the ratio R = PTG/CCA. Serum parathyroid hormone (PTH) is measured 30 minutes post-operatively. A derivation cohort (~120 patients) generates R-based thresholds for in-situ preservation versus autotransplantation, which are prospectively validated in an independent cohort (60 patients) together with decision-curve analysis of clinical net benefit.
Description
Post-thyroidectomy hypocalcemia, reported in 5-30 % of cases, is largely attributable to inaccurate assessment of parathyroid perfusion. Conventional intra-operative judgment-based on gland color, capillary bleeding and pulsation-is subjective and non-quantitative. Near-infrared fluorescence imaging with indocyanine green (ICG) allows real-time visualization of tissue perfusion, yet no widely accepted quantitative threshold exists to guide in-situ preservation versus autotransplantation of parathyroid glands.
The PT-ICG study addresses this gap through:
Target population: Patients undergoing unilateral thyroid lobectomy with planned autotransplantation of the ipsilateral inferior parathyroid gland, isolating evaluation to the superior gland.
Pilot phase (n=30): Optimization confirmed an ICG dose of 25 µg/kg and a 60-120 s acquisition window, with the common carotid artery (CCA) providing a stable reference; well-perfused glands exhibited R values ~0.8-3.0.
Derivation phase (n≈120): The relationship between R and 30-min postoperative PTH is modeled; ROC analysis and the Youden index yield R_low and R_high thresholds, internally validated with bootstrap resampling.
Validation phase (n=60): Prospective application of thresholds documents surgeon decisions, postoperative PTH, and hypocalcemia incidence; predictive performance (AUC, sensitivity, specificity) and net clinical benefit (decision-curve analysis) of the R-based strategy are compared with standard clinical judgment.
Sample size and statistics: With an anticipated correlation r≈0.45, α=0.05 (two-sided) and 90 % power, ≥110 patients are required; allowing 10 % attrition, 120-130 will be enrolled.
Data management and safety: REDCap will support double data entry and monitoring; ICG-related adverse events will be recorded; the study adheres to GCP and the Declaration of Helsinki and has received ethics committee approval.
By establishing actionable ICG fluorescence thresholds and a decision algorithm, the study seeks to provide a quantitative tool for real-time intra-operative perfusion assessment, potentially reducing hypocalcemia and informing future multicenter trials.
full SAP available upon request / will be posted as separate document.
Eligibility
Inclusion Criteria:
- Scheduled for unilateral thyroid lobectomy plus central (VI-level) lymph-node dissection.
- Intra-operative findings consistent with:
- Complete identification and in-situ preservation of the ipsilateral superior parathyroid gland;
- Complete identification and autotransplantation of the ipsilateral inferior parathyroid gland.
- Serum 25-hydroxy-vitamin D ≥ 30 ng/mL at baseline; patients below this level may be enrolled after standardized vitamin-D and calcium supplementation with re-test confirming ≥ 30 ng/mL.
- Surgery and all postoperative follow-up performed at the study center.
- Able and willing to provide written informed consent.
Exclusion Criteria:
- Prior surgery on the thyroid or parathyroid glands.
- Severe hepatic or renal impairment, or other serious metabolic bone disease.
- Pregnancy or lactation.
- Known hypersensitivity to indocyanine green or iodine-containing compounds.
- Inability or unwillingness to comply with postoperative visits and blood testing.