R.m. Schols, MD, N.d. Bouvy, MD, PhD, F.p. Wieringa, PhD, L. Alic, L.p.s. Stassen, MD, PhD
Department of Surgery, Maastricht University Medical Center, Maastricht, The Netherlands; Van ‘t Hoff Program on Medical Photonics, TNO, Eindhoven, The Netherlands
Objective of the technology
Surgery of the thyroid and parathyroid is an advanced procedure, requiring a high level of surgical skills. Differentiating critical tissues takes time and incorporates risks. Iatrogenic injury of the parathyroid glands or the recurrent laryngeal nerve (RLN) is a possible complication that needs to be prevented. Additionally, the visible contrast between thyroid and parathyroid tissue is delicate to observe. This visual task can be even more challenging due to anatomic variations with respect to the localization of the lower parathyroid glands.The aim of this pilot study was to collect in vivo wide-band spectral reflectance-signatures of critical tissue types encountered during thyroid and parathyroid surgery, and to assess the presence of useful spectral distinctive features that might be applied for future devices enabling intraoperative, tissue specific image contrast enhancement. We report the first results for healthy thyroid, parathyroid and RLN.
Description of the technology and method of its use or application
Spectral reflectance-signatures were collected after obtaining institutional review board (IRB) approval and written informed consent. Wide-band spectra (350-1830nm, 1nm resolution) were collected in vivo during thyroid and parathyroid surgery. Subjected to tissue type accessibility, on average 2 tissue types per patient were measured. For each tissue type, 5 spectra were recorded per site (taking around 30s per site), covering 1-2 sites per tissue type. As a first step, mean spectra were calculated for all three measured tissue types. After visually comparing these mean tissue spectra, it was decided to add a secondary step by extracting two spectral signature features for all individual measured sites:
1. Slope within the 650-700nm range;
2. Amplitude gradient between dominant local reflectance minimum and maximum within 1350-1830nm.
In 10 consecutive patients 158 in vivo spectra were recorded on 32 tissue sites. Based on the mean diffuse reflectance spectra for thyroid, parathyroid and RLN (first step of analysis), the estimated features for all three tissue types were plotted (second step of analysis). Significance was tested using a paired Student t-test and by an applied Holm-Bonferoni correction, with the criterion for statistical significance set to p < 0.025. Regarding feature 1 significance was found for parathyroid (p = 0.003) and RLN (p = 0.006). With respect to feature 2, RLN was significant (p = 0.001). Hence, both investigated in vivo spectral features seem to offer added value to distinguish between the studied tissue types.
Conclusions / future directions
To our knowledge, this study is the first to date to investigate in vivo spectral reflectance-signatures of critical tissues encountered during parathyroid and thyroid surgery far beyond the spectral detection boundary of the human eye (the infrared light spectrum). Within the recorded spectral reflectance tissue-signatures in this study we identified two spectral features which may be used for further developments towards intraoperative image contrast enhancement and thereby facilitate the differentiation of tissue structures, either during open or endoscopic surgery.
Session: Poster Presentation
Program Number: ETP013