Author Response: Multidependency of Electroencephalographic and Bispectral Index Reactivity Rule Them Out as Outcome Predictors of Head Trauma Authors: Gayatri Sakrikar, Rohini M Surve, Ramesh Venkatapura, Dhritiman Chakrabarti, Dhaval Shukla, Ravindra Mundlamuri Published: 2026 Dear Editor, We write this letter in response to the Letter to Editor by Josef Finsterer, titled “Multidependency of electroencephalographic and bispectral index reactivity rule them out as outcome predictors of head trauma” by Josef Finsterer. We thank the author for the interest in our study and for the thoughtful comments regarding the interpretation of electroencephalographic reactivity (EEG-R) and bispectral index reactivity (BIS-R) in acute severe traumatic brain injury (sTBI). We appreciate the opportunity to clarify several methodological aspects of our study. We would first like to clarify that the primary objective of our study was to evaluate the incidence of EEG-R and BIS-R and their association with neurological outcomes, rather than treatment success.1 Treatment was not standardized within the study protocol and was administered according to patient condition and at the discretion of the treating physician, in accordance with Brain Trauma Foundation recommendations. Our analysis focused on assessing whether neurological recovery differed between patients with present and absent EEG-R or BIS-R and on evaluating their predictive utility during the acute phase when secondary brain injury processes are ongoing. We agree that intact sensory pathways are essential for reliable assessment of stimulus-induced reactivity. To minimize this confounding, patients with hearing impairment and severe extracranial injuries, including spinal cord injury, were excluded from the study. In addition, all stimuli were applied in a standardized manner, and reproducibility across multiple stimuli was required for EEG-R classification.1 While subclinical peripheral or spinal dysfunction cannot be entirely excluded in critically ill patients, we additionally assessed changes in instantaneous analgesia nociception index values (iANI) during application of painful stimuli and observed statistically significant responses across all four stimuli, supporting adequate nociceptive input. However, this analysis was performed post hoc and was not part of the predefined study objectives. We also agree that TBI represents a heterogeneous spectrum with diverse structural and pathophysiological involvement. To address this, we attempted to maintain clinical uniformity by including only patients with acute sTBI, defined by a Glasgow coma scale score of ≤8, within a narrow post-injury time window. Relevant clinical and imaging characteristics, including injury laterality and pupillary responses, were systematically documented. Wherever feasible, EEG recordings were obtained using all available 16 channels, and reactivity observed in any channel was considered present provided it was reproducible across stimuli.1 While further stratification based on lesion type and anatomical involvement may provide additional insights, the primary objective of our study was to evaluate EEG-R as a functional marker of global brain responsiveness. In this context, EEG-R reflects the integrity of large-scale network interactions and may therefore retain prognostic relevance despite structural heterogeneity. With regard to technical considerations, several precautions were taken to minimize EEG artifacts. Recordings were performed only after impedance verification, and channels with poor signal quality were excluded. Artifact-prone segments were removed during preprocessing, and independent component analysis was employed for quantitative EEG analysis.1 Reactivity assessment was based on relative changes from baseline across multiple channels rather than absolute amplitudes, thereby reducing the influence of focal artifacts related to surgical wounds or skull defects. We agree that systemic factors can influence EEG reactivity, as with any prognostic indicator. However, several steps were taken to minimize such confounding, including exclusion of patients with significant extracranial trauma, sepsis, and major physiological derangements, along with maintaining key biochemical parameters within predefined ranges.1 The relatively high mortality rate observed in our cohort reflects the profile of patients treated at our tertiary referral center. A substantial proportion of patients are referred from peripheral hospitals after initial stabilization or at a later stage of clinical deterioration and often present with higher injury severity, delayed neurosurgical intervention, and refractory intracranial hypertension, all of which contribute to poorer outcomes. We also agree that medications and physiological variables can affect reactivity. To minimize sedative-related confounding, recordings were performed after withholding propofol for at least 2 hours or prior to scheduled opioid dosing, and benzodiazepines were not used. In cases where sedation could not be interrupted, assessment was deferred. Patients with significant metabolic disturbances or sepsis were excluded.1 We acknowledged that evolving intensive care unit (ICU) complications and treatment-related factors may influence outcomes and that serial or continuous monitoring may provide more comprehensive prognostic information than a single assessment and have been mentioned as one of the limitations of the study. Finally, we respectfully disagree with the assertion that the multiple dependencies of EEG-R and BIS-R disqualify them as prognostic tools. Rather, these dependencies emphasize the importance of standardized methodology and careful patient selection, both of which were incorporated into our study design. Our findings demonstrate that absence of EEG-R in the acute phase was highly specific for poor early neurological recovery, supporting its role as an adjunctive prognostic marker. We agree that EEG-R should not be interpreted in isolation but rather as part of a multimodal prognostic framework integrating clinical examination, neuroimaging, and other neurophysiological parameters. Yours sincerely, Gayatri Sakrikar, Rohini M Surve, Ramesh Venkatapura, Dhritiman Chakrabarti, Dhaval Shukla, and Ravindra Mundlamuri Orcid Gayatri Sakrikar https://orcid.org/0000-0002-0480-4570 Rohini M Surve https://orcid.org/0000-0002-5740-7122 Ramesh Venkatapura https://orcid.org/0000-0003-2011-1202 Dhritiman Chakrabarti https://orcid.org/0000-0003-4804-5403 Dhaval Shukla https://orcid.org/0000-0001-7613-1929 Ravindra Mundlamuri https://orcid.org/0000-0002-5360-257X