Association Between Preoperative Delirium and Plasma Biomarkers of Neuropathology Following Acute Hip Fracture: Hip Fracture Inpatient Pathophysiology Study (HIPS) Pilot
Ketura Berry1, Joy Youn1, Julio Rojas-Martinez1, Lawren VandeVrede1, David Castro1, Stephanie Roa Diaz2, Thelma Garcia2, Nida Degesys1, John Boscardin1, Kenneth Covinsky1, Stephanie Rogers1, John Newman2, Bruce Miller1, Adam Boxer1, Vanja Douglas1, Sara LaHue1
1University of California, San Francisco, 2Buck Institute for Research on Aging
Objective:
To investigate the association between preoperative delirium and plasma neuropathology biomarkers.
Background:
Delirium is common in older adults but its association with neuropathology biomarkers in the emergent preoperative setting is less understood.
Design/Methods:
Adults age 65+ hospitalized <72hr after acute hip fracture underwent delirium screening (Confusion Assessment Method) and preoperative blood collection. Plasma neurofilament light chain (NfL), glial fibrillary acidic protein (GFAP) and phosphorylated tau (P-tau217) were measured via Quanterix kits on Simoa HD-X. Cognitive status was determined by preexisting diagnosis and Informant Questionnaire for Cognitive Decline in the Elderly. Descriptive statistics utilized chi-square or Wilcoxon rank-sum. Linear regression explored the association between delirium and biomarkers on preoperative enrollment day.
Results:
Of 16 subjects (ongoing enrollment N=31): mean age 82 years (IQR 73-84), 81% women, 38% had dementia, 50% had delirium preoperatively. 31% reported headstrike with hip fracture; no intracranial pathology on CT. Those with delirium (vs without) had increased prevalence of dementia (80% vs. 18%, p=0.02). Those with delirium (vs without) had higher median NfL [154pg/mL (IQR 31-323) vs 20pg/mL (IQR 17-28), p<0.01], higher median P-tau217 [0.22pg/mL (IQR 0.20-0.31) vs 0.08pg/mL (IQR 0.04-0.12), p=0.01], and trend towards higher median GFAP [352pg/mL (IQR 178-410) vs 215pg/mL (IQR 95-259), p=0.27]. No differences were found in age, sex, or headstrike in those with delirium vs. without. In univariable regressions, NfL was significantly associated with delirium, as was P-tau217. In multivariable regressions adjusting for age and dementia, NfL remained independently associated with delirium (coeff 194pg/mL, 95% CI 74-315, p=0.004); P-tau217 also remained independently associated with delirium (coeff 0.15, 95%CI 0.04-0.27, p=0.01).
Conclusions:
In older adults admitted for emergent hip fracture, preoperative delirium was independently associated with elevated NfL and P-tau217 before surgery, even after controlling for history of dementia. Our findings raise questions about the pathophysiology and time-course of neuronal injury in delirium.