Tissue classification by rapid evaporative ionization mass spectrometry (REIMS): comparison between a diathermic knife and CO2 laser sampling on classification performance

Abstract

The increasing need for rapid, in situ, and robust tissue profiling approaches in the context of intraoperative diagnostics has led to the development of a large number of ambient ionization-based surface sampling strategies. This paper compares the performances of a diathermic knife and a CO₂ laser handpiece, both clinically approved, coupled to a rapid evaporative ionization mass spectrometry (REIMS) source for quasi-instantaneous tissue classification. Several fresh meat samples (muscle, liver, bone, bone marrow, cartilage, skin, fat) were obtained from different animals. Overall, the laser produced cleaner cuts and more reproducible and higher spectral quality signals when compared with the diathermic knife (CV laser = 9-12%, CV diathermic = 14-23%). The molecular profiles were subsequently entered into a database and PCA/LDA classification/prediction models were built to assess if the data generated with one sampling modality can be employed to classify the data generated with the other handpiece. We demonstrate that the correct classification rate of the models increases (+ 25%) with the introduction of a model based on peak lists that are tissue-specific and common to the two handpieces, compared with considering solely the whole molecular profile. This renders it possible to use a unique and universal database for quasi-instantaneous tissue recognition which would provide similar classification results independent of the handpiece used. Furthermore, the laser was able to generate aerosols rich in lipids from hard tissues such as bone, bone marrow, and cartilage. Combined, these results demonstrate that REIMS is a valuable and versatile tool for instantaneous identification/classification of hard tissue and coupling to different aerosol-generating handpieces expands its field of application. Graphical abstract.

Keywords: Bioanalytical methods; Laser ablation; Lipids; Mass spectrometry; REIMS; Tissue analysis.

Graphical abstract

Fig. 1

Experimental workflow from tissue sampling to the creation of tissue classification models: Every tissue (chicken [Ch.], rabbit [Rb.], duck [Du.], turkey [Tu.], cow [Co.], and calf [Ca.]) was sampled and analyzed with both the CO₂ laser [a] and the electrocautery knife [b]. For safety reasons, the laser handpiece was mounted on a home-built mechanic arm [c] maintained in a fixed position during laser ablation. Both handpieces were connected to the REIMS source and the data collected from vapors were acquired in negative ionization mode [d]. After basic data preprocessing, several PCA-LDA predictive models were created based on the overall molecular (lipids) fingerprint [e]. From the loading plots of each model, tissue-specific peaks were selected to create peak lists common to the two handpieces. This peak list was further used to build a more targeted model to improve the classification rate of our statistical models that are now able to classify tissues independently from the handpiece used for sampling

Fig. 2

Macroscopic [a, c] and microscopic (H&E stained) [b, d] visualization of the tissue damages generated after laser [a, b] and diathermic [c, d] samplings. The laser generated a smaller tissue damage compared with the diathermic knife (laser ⌀ = 0.39 mm, diathermic ⌀ = 0.79 mm wide/3.65 mm long). The laser also produced a more reproducible signal, reduced spectral noise, and more lipid clusters (in the region m/z 1200–1500) [e1–e2]. FA, fatty acids; GP, glycerophospholipids

Fig. 3

A PCA-LDA predictive model used for the creation of inclusion and exclusion peak lists. [a] Score plot from a model including cow liver and chicken liver analyzed with the two handpieces. From the loading plot, peaks able to discriminate between the two tissue types [b] were selected as an “inclusion list” (referred to as “Inc.”). Peaks able to discriminate between the two handpieces [c] were selected as an “exclusion list” (referred to as “Excl.”). The two peak lists were compared, and the peaks included in the exclusion list were removed from the inclusion list in order to create tissue-specific peak lists (tissue-specific peak list = Inc. − Excl.)

Fig. 4

Schematic interpretation of the influence of the data variation in the creation of a predictive model. The data not included in the ellipse (represented by the light-colored ovals) are those which returned as outliers by the cross-validation process. This illustrates why the use of a more precise model leads to more outliers when classifying a subsequent dataset which inherently comprises more variance

Fig. 5

Sampling of hard tissue with the laser-REIMS system. With the CO₂ laser, it was possible to generate a signal from bone [a], bone marrow [b], and cartilage [c]. It is not possible to generate aerosols and consequently a signal while using the diathermic knife due to the hard nature of the bone