Clinical Manifestations.

Background: Postoperative complications of major surgical interventions include delirium. Delirium is a risk factor for dementia, and in some cases, may signal underlying neuropathological processes. Cognitive tests that accurately predict post-operative outcomes could identify patients with cognitive vulnerabilities who may benefit from preoperative counseling and postoperative interventions. Unfortunately, in-person presurgical cognitive evaluations are often not available, but digital technologies may improve access. In this pilot study, we evaluate the feasibility of utilizing a smartphone app to conduct preoperative remote cognitive assessment in patients undergoing spine surgery. We also evaluate the relationship of app-based assessments with subjective cognitive concerns.

Method: We enrolled 30 older adults (≥ 65 years old) who were scheduled for elective spinal surgery anticipated to last at least three hours. On the ALLFTD Mobile App, participants completed a battery of cognitive tests, a measure of subjective cognitive concerns (12-item self-reported everyday cognition scale, Ecog-12), and a user experience survey. A subset (N = 12) had a study partner complete the 41-item Ecog, reporting on observed changes in cognitive function over a 10-year period prior to the surgery. Linear regressions adjusted for age evaluated the associations between subjective cognitive concerns and the objective app-based outcomes (Flanker task and a composite of several measures).

Result: Average age of participants was 71.2 years (SD = 4.4); 46.1% were women and 75% identified as White. Most (96%) participants indicated that the time required to complete the app assessments was acceptable, 77% indicated that the app instructions were clear, and all participants indicated that the text on the app was large enough to read comfortably. Participants' subjective cognitive concerns (ECog-12) were significantly associated with a composite of app tasks (β = -0.48, p = 0.01). The executive functioning domain of the study partner Ecog was significantly associated with patient performance on an app-based executive functioning task (β = 0.75, p = 0.007).

Conclusion: These findings support the feasibility of remote preoperative assessments in older adults using smartphones. The study also provides preliminary support that these measures can objectively quantify self- and partner-reported cognitive concerns. Future work will explore associations with postoperative outcomes.