Extension of the Dutch Linguistic Intraoperative Protocol: New Standardized Items and Tasks for Language Monitoring in Awake Glioma Surgery for Mild, Moderate and Severe Aphasia

BACKGROUND
Awake surgery with intraoperative language monitoring is the standard treatment for low-grade glioma patients, and also seems to result in better surgical outcomes in glioblastoma patients. The Dutch Linguistic Intraoperative Protocol (DuLIP) is a validated tool for intraoperative language monitoring and is available in several languages. However, in clinical practice, addition of tasks and items appeared to be necessary to 1) enlarge item and task variation 2) add less complex tasks for patients with preoperative severe aphasia. Therefore, we aimed to develop an extended version of DuLIP (DuLIP-Extended).

MATERIAL AND METHODS
We enlarged and developed 24 tasks. 1) Extra items for word repetition and the original semantic tasks. 2) Additional tasks for phonology (reading aloud sentences, judgment of onset phonemes, rhyming judgment, rhyming), semantics (word pairs) and syntax (syntactically induced sentence completion, converting sentences from the present tense to the past tense). 3) Tasks for severely aphasic patients for phonology (rhyming with cue), semantics (matching pictures to categories or to written words), and object naming of high-frequent words. DuLIP-Extended was administered in 160 healthy native speakers of Dutch with a median age of 53 years (range 18-83 years). The accuracy rate was calculated per item and per task. Differences on task performance between groups of age (</≥ 55 years) and education (</≥ bachelor) were investigated with ANOVAs.

RESULTS
All tasks had an accuracy rate of ≥ 90%. Seven items were removed due to too low accuracy rate of ≥ 80% and 20 items were removed which evoked emotional, stereotype or ambiguous responses. Younger participants (< 55 years) performed significantly better (p <.05) than older participants (≥ 55 years) on 14/24 tasks such as rhyming and semantic association. Higher educated participants (≥ bachelor) performed significantly better (p <.05) than lower educated participants (< bachelor) on 5/24 tasks, such as repetition of compound words and converting sentences to past tense. Cut-offs were created per task, based on the significant predictors.

CONCLUSION
DuLIP-Extended appeared to be feasible in a large group of healthy participants. Like the original DuLIP, the new items and tasks have a high accuracy rate and as expected effects of age and education were found. DuLIP-Extended will enable to select from a larger variety of items and tasks at different linguistic levels for glioma patients with mild to severe aphasia. This allows for a more inclusive neuro-oncological care in the pre-, intra- and postoperative phase. In future research, validation and application of DuLIP-Extended in glioma patients, according to cortico-subcortical areas will be performed. Adaptation to other languages will follow.