We have used AMIGO for the localization of parathyroid glands during parathyroidectomy and recurrent laryngeal nerves during thyroidectomy. Finding parathyroid glands can be challenging, even for the experienced endocrine surgeon, particularly when the glands are ectopic or are present in a re-operative setting. The current methods for localization of parathyroid adenomas rely on preoperative localization with ultrasound and sestamibi; however, these tests are not totally reliable and there are no useful intra-operative imaging methods for detecting parathyroid adenomas. The AMIGO-guided surgery allows the surgeon to focus in on the target parathyroid gland using the navigation system. By comparing the virtual space model of the tumor to the actual dissection, the surgeon can assess the location of the tumor and the extent to which the resection is complete.

Currently, there are no intra-operative imaging techniques that can aid in the localization of a parathyroid tumor. The surgeon must rely on expertise and judgment when the preoperative localization fails or is discordant. The AMIGO-guided parathyroid surgery provides real-time intraoperative feedback that assesses the location and degree of resection of parathyroid tumors that has not heretofore been possible.

We have attempted to map the recurrent laryngeal nerve -- a nerve that controls the voice -- during thyroid surgery. There are currently widely used nerve stimulators for detecting these nerves, but none is able to image them and define their course preoperatively. Using this innovation, we would be able to map this important nerve during thyroid surgery for large goiters or invasive cancers. This technology has the potential for becoming useful in the future, for re-operative parathyroid surgery, and in cases of invasive/anaplastic thyroid cancer in which the nerves can be mapped out in advance.

The case below illustrates the workflow for parathyroidectomy in AMIGO. A 28-year-old male presented with hypercalcemia and multiple symptoms associated with hyperparathyroidism. A diagnostic workup confirmed autonomous hypersecretion of the parathyroid hormone. Preoperative Tc-Sestamibi imaging revealed a focus of increased radiotracer uptake deep within the central neck in the tracheoesophageal groove. The small and deep-seated tumor would have required a larger incision and more extensive field of dissection without image-guidance. The patient underwent focused left parathyroidectomy in AMIGO.

Parathyroidectomy Workflow in AMIGO

Diagnostic CT Imaging showing the parathyroid adenoma.
Diagnostic Tc-Sestamibi imaging showing an abnormal tracer uptake
Diagnostic US imaging showing the adenoma
Surgical Planning: 3D model generated from preoperative CT using 3D Slicer software - anterior view
3D model using 3D Slicer software - left view
3D model using 3D Slicer software - posterior view
The patient was placed on the operating table in the surgical position and intraoperative MRI was obtained. Using these intraoperative images, three dimensional models of the anatomy were created and provided to the navigation system.
A Bovi pointer was instrumented with a position/orientation sensor to localize the different structures within the neck and guide the surgeon to the adenoma. The use of the navigation system allowed for the procedure to be completed promptly through a tiny incision with minimal dissection. The navigation system display showing the instrument (red cylinder) with respect to the neck anatomy (skin-peach, trachea-aqua, thyroid-yellow, tumor-green (deformed) and blue (undeformed))
AMIGO setup showing the Slicer navigation display on the monitor (indicated by red arrow)

Select Publications

Calligaris D, Feldman DR, Norton I, Olubiyi O, Changelian AN, Machaidze R, Vestal ML, Laws ER, Dunn IF, Santagata S, et al. MALDI Mass Spectrometry Imaging Analysis of Pituitary Adenomas for Near-real-time Tumor Delineation. Proc Natl Acad Sci U S A. 2015;112 (32) :9978-83.Abstract

We present a proof of concept study designed to support the clinical development of mass spectrometry imaging (MSI) for the detection of pituitary tumors during surgery. We analyzed by matrix-assisted laser desorption/ionization (MALDI) MSI six nonpathological (NP) human pituitary glands and 45 hormone secreting and nonsecreting (NS) human pituitary adenomas. We show that the distribution of pituitary hormones such as prolactin (PRL), growth hormone (GH), adrenocorticotropic hormone (ACTH), and thyroid stimulating hormone (TSH) in both normal and tumor tissues can be assessed by using this approach. The presence of most of the pituitary hormones was confirmed by using MS/MS and pseudo-MS/MS methods, and subtyping of pituitary adenomas was performed by using principal component analysis (PCA) and support vector machine (SVM). Our proof of concept study demonstrates that MALDI MSI could be used to directly detect excessive hormonal production from functional pituitary adenomas and generally classify pituitary adenomas by using statistical and machine learning analyses. The tissue characterization can be completed in fewer than 30 min and could therefore be applied for the near-real-time detection and delineation of pituitary tumors for intraoperative surgical decision-making.