Glioblastoma Pseudoprogression Discrimination Using Multiparametric Magnetic Resonance Imaging, Principal Component Analysis, and Supervised and Unsupervised Machine Learning.

BACKGROUND: One of the most frequent phenomena in the follow-up of glioblastoma is pseudoprogression, present in up to half of cases. The clinical usefulness of discriminating this phenomenon through magnetic resonance imaging and nuclear medicine has not yet been standardized; in this study, we used machine learning on multiparametric magnetic resonance imaging to explore discriminators of this phenomenon. METHODS: For the study, 30 patients diagnosed with IDH wild-type glioblastoma operated on at both study centers in 2011-2020 were selected; 15 patients corresponded to early tumor progression and 15 patients to pseudoprogression. Using unsupervised learning, the number of clusters and tumor segmentation was recorded using gap-stat and k-means method, adjusting to voxel adjacency. In a second phase, a class prediction was carried out with a multinomial logistic regression supervised learning method; the outcome variables were the percentage of assignment, class overrepresentation, and degree of voxel adjacency. RESULTS: Unsupervised learning of the tumor in its diagnosis shows up to 14 well-differentiated tumor areas. In the supervised learning phase, there is a higher percentage of assigned classes (P < 0.01), less overrepresentation of classes (P < 0.01), and greater adjacency (55% vs. 33%) in cases of true tumor progression compared with pseudoprogression. CONCLUSIONS: True tumor progression preserves the multidimensional characteristics of the basal tumor at the voxel and region of interest level, resulting in a characteristic differential pattern when supervised learning is used.

Target location after deep cerebral biopsies using low-volume air injection in 75 patients. Results and technical note.

BACKGROUND: Stereotactic biopsy is a minimally invasive technique that allows brain tissue samples to be obtained with low risk. Classically, different techniques have been used to identify the biopsy site after surgery. OBJECTIVE: To describe a technique to identify the precise location of the target in the postoperative CT scan using the injection of a low volume of air into the biopsy cannula. METHODS: Seventy-five biopsies were performed in 65 adults and 10 children (40 males and 35 females, median age 51 years). Frame-based biopsy was performed in 46 patients, while frameless biopsy was performed in the remaining 29 patients. In both systems, after brain specimens had been collected and with the biopsy needle tip in the center of the target, a small volume of air (median 0.7 cm3) was injected into the site. RESULTS: A follow-up CT scan was performed in all patients. Intracranial air in the selected target was present in 69 patients (92%). No air was observed in two patients (air volume administered in these 2 cases was below 0.7 cm3), while in the remaining four patients blood content was observed in the target. The diagnostic yield in this series was 97.3%. No complications were found to be associated with intracranial air injection in any of the 75 patients who underwent this procedure. CONCLUSIONS: The air-injection maneuver proposed for use in stereotactic biopsies of intracranial mass lesions is a safe and reliable technique that allows the exact biopsy site to be located without any related complications.

Idiopathic normal pressure hydrocephalus: results of a prospective cohort of 236 shunted patients.

AIM: To describe the outcomes and complication rates in 236 patients with idiopathic normal pressure hydrocephalus (INPH) after treatment. PATIENTS AND METHODS: Among a cohort of 257 patients with suspected INPH, 244 were shunted and 236 were followed up at 6 months after shunting (145 men [61.4%] and 91 women [38.6%] with a median age of 75 years). The study protocol of these patients included clinical, radiological, neuropsychological and functional assessment. The decision to shunt patients was based on continuous intracranial pressure monitoring and CSF dynamics studies. A differential low-pressure valve system, always combined with a gravity compensating device, was implanted in 99% of the patients. RESULTS: After shunting, 89.9% of the patients showed clinical improvement (gait improved in 79.3% of patients, sphincter control in 82.4%, and dementia in 63.7%). Two patients (0.8%) died. Early postsurgical complications were found in 13 of the 244 shunted patients (5.3%). Six months after shunting, the follow-up CT showed asymptomatic hygromas in 8 of the 236 (3.4%). Additional postsurgical complications were found in 7 patients (3%), consisting of 6 subdural hematomas (3 acute and 3 chronic) and 1 distal catheter infection. CONCLUSIONS: Currently, a high percentage of patients with INPH can improve after shunting, with early and late complication rates of less than 12%.