Impact of diagnostic investigations in the management of CAR T-cell-associated neurotoxicity.

ABSTRACT: International guidelines regarding the management of immune effector cell-associated neurotoxicity syndrome (ICANS) recommend several diagnostic investigations, including magnetic resonance imaging (MRI), lumbar puncture (LP), and electroencephalogram (EEG) based on ICANS grade. However, the impact of these investigations has not yet been evaluated. Here, we aimed to describe the role of MRI, LP, and EEG in the management of ICANS in a cohort of real-life patients treated with chimeric antigen receptor (CAR) T cells at the University Hospital of Rennes, France. Between August 2018 and January 2023, a total of 190 consecutive patients were treated with CAR T cells. Among those, 91 (48%) developed ICANS. MRI was performed in 71 patients (78%) with ICANS, with a therapeutic impact in 4% of patients, despite frequent abnormal findings. LP was performed in 43 patients (47%), which led to preemptive antimicrobial agents in 7% of patients, although no infection was eventually detected. Systematic EEG was performed in 51 patients (56%), which led to therapeutic modifications in 16% of patients. Our study shows that EEG is the diagnostic investigation with the greatest therapeutic impact, whereas MRI and LP appear to have a limited therapeutic impact. Our results emphasize the role of EEG in the current guidelines but question the need for systematic MRI and LP, which might be left to the discretion of the treating physician.

Moderate-to-severe ARDS: COVID-19 patients compared to influenza patients for ventilator parameters and mortality.

Background: This study aimed to compare ventilatory parameters recorded in the first days of acute respiratory distress syndrome (ARDS) and mortality at day 60 between coronavirus disease 2019 (COVID-19) and influenza ARDS patients with arterial oxygen tension (P aO2 )/inspiratory oxygen fraction (F IO2 ) ≤150 mmHg. Methods: We compared 244 COVID-19 ARDS patients with 106 influenza ARDS patients. Driving pressure, respiratory system compliance (C rs), ventilator ratio, corrected minute ventilation (V'Ecorr) and surrogate of mechanical power (index=(4×driving pressure)+respiratory rate) were calculated from day 1 to day 5 of ARDS. A propensity score analysis and a principal component analysis (PCA) were performed. Results: On day 1 of ARDS, COVID-19 patients had significantly higher P aO2 /F IO2 (median (interquartile range) 97 (79-129.2) versus 83 (62.2-114) mmHg; p=0.001), and lower driving pressure (13.0 (11.0-16.0) versus 14.0 (12.0-16.7) cmH2O; p=0.01), ventilatory ratio (2.08 (1.73-2.49 versus 2.52 (1.97-3.03); p<0.001), V'Ecorr (12.7 (10.2-14.9) versus 14.9 (11.6-18.6) L·min-1; p<0.001) and index (80 (70-89) versus 84 (75-94); p=0.004). PCA demonstrated an important overlap of ventilatory parameters recorded on day 1 between the two groups. From day 1 to day 5, repeated values of P aO2 /F IO2 , arterial carbon dioxide tension, ventilatory ratio and V'Ecorr differed significantly between influenza and COVID-19 patients in the unmatched and matched populations. Mortality at day 60 did not differ significantly after matching (29% versus 21.7%; p=0.43). Conclusions: Ventilation was more impaired in influenza than in COVID-19 ARDS patients on the first day of ARDS with an important overlap of values. However, mortality at day 60 did not differ significantly in the matched population.

Improving Academic Performance in Medical Students Using Immersive Virtual Patient Simulation: A Randomized Controlled Trial.

OBJECTIVES: To determine whether immersive virtual patient simulation (IVPS) on the MedicActiv platform is influential in improving student academic performance in module validation tests. DESIGN: In this prospective randomized controlled study a comparison was made between IVPS training combined with regular faculty courses versus courses alone. The primary endpoint was module validation grades. Secondary endpoints were satisfaction scores (overall interest, ergonomics, realism, immersion, and training efficiency). SETTING: Angers School of Medicine, France. PARTICIPANTS: 2018 to 2019 class of fourth-year students, included on a voluntary basis. In the first semester 51 students were included, of whom 13 were excluded for lack of compliance (6 and 7 from the IVPS and control groups, respectively). In the second semester we included 57 students, of whom 10 were excluded for lack of compliance (2 and 8 from the IVPS and control groups, respectively). RESULTS: Mean age was 21 years (±0.8). There were 85 female and 23 male students. In the first semester mean grades were 13.4 ± 1.6 versus 11/.9 ± 2.4 in the IVPS and control groups, respectively (p = 0.038). In the second semester mean grades were 15.3 ± 2.5 versus 11.9 ± 3.6 in the IVPS and control groups, respectively p < 0.001. The entire study population was pooled (n = 85): mean grades were 14.5 ± 2.4 versus 11.9 ± 3 in the IVPS group and the control group, respectively, p < 0.001. The satisfaction questionnaire response rate was 54% (46/85). Score percentages ≥4 regarding overall interest, ergonomics, realism, immersion and training efficiency were 89%, 85%, 100%, 93%, and 93% respectively. CONCLUSIONS: Complementing conventional university education with simulation of virtual consultation cases on the MedicActiv platform improved student academic performance as compared with students studying regular courses. Students reported high levels of satisfaction with overall interest, ergonomics, realism, immersion and training efficiency on the MedicActiv platform.