Autofluorescence in Surgery for Primary Hyperparathyroidism

Parathyroidectomy (PTX) is the only curative treatment for primary hyperparathyroidism (PHPT).1 Prior to surgery, imaging is performed to determine the anatomical location of pathological parathyroid gland (PTGs). Usually two independent localizing imaging methods are used, e.g. sestamibi scintigraphy and neck ultrasound. Additional imaging modalities may be added, e.g. PET/CT. In case of concordant imaging results, a selective/focused PTX may be performed with removal of the identified pathological gland. If imaging is negative or non-concordant, an exploratory parathyroidectomy (ePTX) is performed, either as a bilateral four-gland exploration, or as a unilateral exploration in case of suspicion to one side of the neck.2 In exploratory PTX, the surgeon must use experience to visually identify both normal and pathological PTGs, which may be very challenging. Identification of normal PTGs will usually aid in determining the likely location of missing pathological PTGs. Failure to localize and remove pathological PTGs will result in a failed operation with persistent disease.

Results of a large European survey indicate that approximately 1% of patients undergo re- operation shortly after the first surgery, and 4,3% had persistent hypercalcemia at first follow-up.3 Recently, near-infrared autofluorescence imaging (AF) has been used as an intraoperative adjunct to localize PTGs. This may assist the surgeon in identifying both normal and pathological PTGs. Potentially, the use of AF could the result in higher probability of cure from PTX, as well as shorter operating time and possibly less extensive surgery and less complications.

1. Study design Randomized, controlled trial. Patients will be randomized 1:1 to either control or AF group.
2. Study setting and practical conduct The trial is to be conducted at Department of Otorhinolaryngology - Head \& Neck Surgery, Aarhus University Hospital (AUH). Here, approximately 300 patients undergo surgery for PHPT annually, of which approximately half undergo an exploratory PTX, and would be eligible for this study. Thus, there is a steady high flow of eligible patients, and the department as well as the treating surgeons are highly qualified to undertake this study. Autofluorescence imaging with the EleVision IR system is available at the department, and is used ad hoc for thyroid and parathyroid surgery. Patient flow Patients referred for treatment of PHPT is seen at the parathyroid clinic at AUH in collaboration between a surgeon and an endocrinologist, and will be considered for inclusion in the trial. The investigators aim to include 100 patients, which should be achievable in approximately two years. Randomization Following patient consent, an electronic case file will be created in the REDCap electronic data capture tool. Randomization is performed electronically in the dedicated REDCap randomization module. Randomization will be stratified according to high confidence finding on imaging or not (Confidence grade 3 vs. 0-2). Intervention Patients randomized to the experimental group will have surgery performed in the exact same manner as in the control group. Patients will be operated with a standard exploratory PTX, with visual identification of PTGs. Unilateral or bilateral parathyroid exploration will be determined by the surgeon, depending on preoperative imaging and intraoperative findings. Usually, the procedure will be terminated when intraoperative parathyroid hormone (PTH) measurement shows a >50% decline, after removal of suspected pathological PTGs. In the experimental group, the surgeon will use the EleVision IR camera system (Medtronic, USA) to visualize PTGs during surgery. The surgical field will be visualized with AF as a minimum of two times during surgery on each side of the neck: First upon exposing the undersurface of the thyroid gland, and secondly before removal of a pathological PTG. Also, removed PTGs will be examined ex vivo to document AF pattern.
3. Deviation from standard treatment. All patients, whether randomized to standard or experimental treatment arm, will receive treatment according to national clinical guidelines. The extent of surgery will be exactly the same in either group, and will not deviate from standard treatment.

Standard surgery for PHPT is performed by open cervical approach through a neck skin incision.

1. Statistical considerations. Sample size Based on data from the last 5 years of surgery, the mean operating time for an ePTX is 102 minutes. The investigators estimate that if the need for repeat intraoperative exploration after ioPTH measurement is reduced from 3/10 to 1/10, a reduction in mean operating time from 98 minutes to 83 minutes would be clinically significant, and statistically significant at α=0.05 with a power of β=0.8 with inclusion of 90 patients.2 To allow for dropouts, the investigators aim to include 100 patients, which should be achievable in approximately two years.

Interim analysis The investigators plan to perform an interim analysis after inclusion of 50 patients with available data on the primary endpoint. The trial may be terminated if the investigators are able to show that it would be highly unlikely to reach the primary endpoint based on a non-inferiority analysis. This may be due to no effect of the intervention, or a lower than expected operating time in the control group. Both scenarios would make it highly unlikely to reach the primary endpoint of the study.

Statistical analysis plan For the primary endpoint, the time from skin incision to termination of the operation is recorded. Reduction of operating time and its 95% confidence interval is calculated and compared with a t- test. In case of non-normal distribution, a rank-sum test will be used instead. For secondary endpoints, a similar analysis will be performed for continuous data (ioPTH reduction; number of ioPTH measurements; number of PTGs identified), and for analysis of categorical data (proportion with biochemical cure; extent of surgery; complications) a chi2-test will be used. α=0.05 will be considered statistically significant.

Patient characteristics, including demographics, disease status and treatment will be characterized using descriptive statistics only. Exploratory sub-group analyses may be performed.