Mackenzie-Phelan, R (2024) Optimising perioperative language assessment in awake craniotomy: picture naming with nouns and finite verbs in the past and present tense. Doctoral thesis, Liverpool John Moores University.

Preview

Text Mackenzie-Phelan_Thesis_2024.pdf - Published Version Available under License Creative Commons Attribution Non-commercial. Download (4MB) | Preview

Abstract

Direct electrical stimulation to map neural language functions during awake craniotomy for low-grade gliomas (brain tumours) is the gold-standard neurosurgical approach. This technique, known as cortical mapping, involves delivering inhibitory electrical currents to brain regions while language functions are assessed through simultaneous neuropsychological testing. Current intraoperative language assessment protocols, however, lack standardisation and rigour. Basic tasks such as counting, object naming, and reading offer limited scope to assess more intricate linguistic components such as grammatical processing (e.g., verbs vs. nouns). This presents significant limitations to intraoperative language mapping that may compromise preserving patients’ language functions postoperatively. Moreover, insensitivity in testing also extends to pre- and post-operative clinical assessment. Standard aphasia assessment is designed predominantly for stroke patients and recent studies demonstrate its limited capacity to capture subtle impairments in the preoperative glioma language profile (e.g., accuracy vs. reaction time). The aim of the present thesis was to address this issue and improve postoperative language outcomes for low-grade glioma patients undergoing awake craniotomy. Through a systematic review of the neurosurgical literature, Study 1 (Chapter 3) aimed to synthesise data from brain stimulation mapping studies of different cognitive and linguistic tasks used during awake craniotomy. This provided an improved understanding of the brain areas successfully mapped with different tasks to support the development and implementation of comprehensive protocols for optimising cognition and language mapping; both among the clinical and research community. For the first time within the UK NHS, a standardised English version of a new linguistic protocol already in practice around Europe, was adopted and trialled for pre-, intra and post-operative language assessment (Study 2, Chapter 4). The Verb and Noun Test for perioperative testing (VAN-POP) consists of object naming and action naming with finite verbs (ANFV) which come together to assess the complex linguistic components (semantic, phonological and grammatical processes) involved in sentence production. However, unlike versions developed in other languages which only assess present tense finite verb production, the English version includes an additional subset assessing finite verb production in the past tense. Therefore, this study novelly implements a three-task approach of object naming and ANFV in the past and present tense. The VAN-POP successfully mapped and monitored language in four patients with suspected low-grade gliomas (frontal, parietal, temporal, and fronto-temporal). Additionally, these tasks enabled the detection of some novel grammatical interferences relating to tense and inflection that have not yet been reported in the awake neurosurgical literature. Study 3 (Chapter 5) adapted the VAN-POP into speeded naming tasks to assess preoperative baseline and short and longer-term postoperative changes in language function. Accuracy and reaction time data were collected in three patients recruited for Study 2 (Chapter 4) at preoperative, postoperative (1-month) and follow-up timepoints (3-month) and compared with healthy controls. Variable performance (improvements and declines) was observed for different tasks in patients over the postoperative course compared to baseline and control performance. Crucially, reaction times, which often go unmeasured clinically, captured the patients’ language impairment better than accuracy at all testing stages. At follow-up, all patients were found to have impaired lexical retrieval speed on at least two tasks either in relation to controls or their preoperative baseline. However, none of the patients were significantly impaired with regards to accuracy. This finding suggests that both accuracy and reaction time measures are important in obtaining a comprehensive understanding of impairment within the glioma language profile. Moreover, the finding which was most consistent across all patients was a greater impairment in retrieval for past tense ANFV compared to present tense ANFV. This novel finding may be further understood within the context of time-reference theories and suggests that more specific testing of verb morphology in glioma patients is warranted. This research has expanded on previous studies that have established the application of object and action naming to optimise intraoperative and perioperative assessment in frontal, temporal, and parietal glioma. Importantly, in line with other recent research, slower processing speed appears to be a more central characteristic of the glioma language profile; with these subtler, yet salient, impairments shown to extend and further decline postoperatively. Collectively, the findings of this thesis question the status quo of neurosurgical practice and neuropsychological testing in terms of the sensitivity and scope of intraoperative and perioperative tasks. As demonstrated, incorporating rigorous and comprehensive linguistic testing such as VAN-POP, in the pre-, intra, and post-operative period is crucial for assessing the intricacies of the glioma language profile and in turn, maximising postoperative neuropsychological outcomes.

Item Type:	Thesis (Doctoral)
Uncontrolled Keywords:	Awake craniotomy; Awake neurosurgery; Glioma; Brain tumours; Language mapping; Action naming with finite verbs; Aphasia
Subjects:	B Philosophy. Psychology. Religion > BF Psychology
Divisions:	Psychology (from Sep 2019)
SWORD Depositor:	A Symplectic
Date Deposited:	10 Jun 2024 14:00
Last Modified:	10 Jun 2024 14:00
DOI or ID number:	10.24377/LJMU.t.00023366
Supervisors:	Roberts, DJ, McGlone, FP and Brooks, SJ
URI:	https://researchonline.ljmu.ac.uk/id/eprint/23366

View Item