Intraoperative in vivo confocal endomicroscopy of the glioma margin: performance assessment of image interpretation by neurosurgeon users

Abstract

Objectives: Confocal laser endomicroscopy (CLE) is an intraoperative real-time cellular resolution imaging technology that images brain tumor histoarchitecture. Previously, we demonstrated that CLE images may be interpreted by neuropathologists to determine the presence of tumor infiltration at glioma margins. In this study, we assessed neurosurgeons' ability to interpret CLE images from glioma margins and compared their assessments to those of neuropathologists.

Methods: In vivo CLE images acquired at the glioma margins that were previously reviewed by CLE-experienced neuropathologists were interpreted by four CLE-experienced neurosurgeons. A numerical scoring system from 0 to 5 and a dichotomous scoring system based on pathological features were used. Scores from assessments of hematoxylin and eosin (H&E)-stained sections and CLE images by neuropathologists from a previous study were used for comparison. Neurosurgeons' scores were compared to the H&E findings. The inter-rater agreement and diagnostic performance based on neurosurgeons' scores were calculated. The concordance between dichotomous and numerical scores was determined.

Results: In all, 4275 images from 56 glioma margin regions of interest (ROIs) were included in the analysis. With the numerical scoring system, the inter-rater agreement for neurosurgeons interpreting CLE images was moderate for all ROIs (mean agreement, 61%), which was significantly better than the inter-rater agreement for the neuropathologists (mean agreement, 48%) (p < 0.01). The inter-rater agreement for neurosurgeons using the dichotomous scoring system was 83%. The concordance between the numerical and dichotomous scoring systems was 93%. The overall sensitivity, specificity, positive predictive value, and negative predictive value were 78%, 32%, 62%, and 50%, respectively, using the numerical scoring system and 80%, 27%, 61%, and 48%, respectively, using the dichotomous scoring system. No statistically significant differences in diagnostic performance were found between the neurosurgeons and neuropathologists.

Conclusion: Neurosurgeons' performance in interpreting CLE images was comparable to that of neuropathologists. These results suggest that CLE could be used as an intraoperative guidance tool with neurosurgeons interpreting the images with or without assistance of the neuropathologists. The dichotomous scoring system is robust yet simple and may streamline rapid, simultaneous interpretation of CLE images during imaging.

Keywords: brain tumor; confocal laser endomicroscopy; fluorescein sodium; glioma; intraoperative imaging; tumor margin.

Figure 1

Box and whisker plots showing individual scoring using the numerical scale by (A) each neurosurgeon (NS) and (B) each neuropathologist (NP). The boxes extend from the 25th to the 75th percentiles. The lines in the middle of the boxes represent the median of the scores and may coincide with the 25th or the 75th percentiles. The whiskers show the maximum and minimum scores given by the rater. Used with permission from Barrow Neurological Institute, Phoenix, Arizona.

Figure 2

Representative images for CLE image scores in the numerical scoring system. (A) Score of 1: CLE image with normal cellularity. (B) Score of 2: CLE image with slightly elevated cellularity. This finding is not sufficient to confirm tumor infiltration. (C) Score of 3: Significantly elevated cellularity, atypia, or distorted architecture, most likely due to tumor infiltration. (D) Score of 4: Marked tumor infiltration is obvious. (E) Score of 5: Solid tumor. No ROI was unanimously scored 0, which indicates faint signal and no obvious identifiable structures on the CLE images. Used with permission from Barrow Neurological Institute, Phoenix, Arizona.

Figure 3

Representative CLE images from 3 ROIs that had a discrepancy of interpretation between neurosurgeons and neuropathologists. In ROI 1, (A) one sample showed normal cellularity and no pathological features, suggesting nontumorous tissue, and (B) a second sample showed nests of cells that could be interpreted as infiltrating tumor cells or clumps of erythrocytes. In ROI 2, (C) one sample showed scattered cells (likely erythrocytes) and several small vessels over a relatively normocellular background and (D) a second sample showed highly cellular density, but it was difficult to determine whether the cells represented tumor cells or erythrocytes. In ROI 3, (E) one sample showed a relatively normal cellular density and morphology, with cord-like structures (arrows) likely representing axons; and (F) a second image from the same Z-stack sequence showed some cells with large atypical nuclei (arrowheads) that were suspicious for tumor cells. Used with permission from Barrow Neurological Institute, Phoenix, Arizona.