Preoperative serum amyloid A: a practical window into neuroinflammatory vulnerability for postoperative delirium Authors: Jae Hoon Lee Published: 2026 Postoperative delirium (POD) remains one of the most consequential complications of major surgery in older adults [1]. After hip surgery in particular it is strikingly common; a recent meta-analysis of nearly 19 000 patients estimated a pooled prevalence of approximately 23% and confirmed advanced age, a high American Society of Anesthesiologists physical status score, dementia, diabetes, hypertension, general anesthesia, male sex, and a prior history of delirium as robust risk factors [2]. POD is far from a transient inconvenience; it is associated with prolonged hospitalization, functional decline, and an elevated long-term risk of dementia and sustained cognitive deterioration [3]. Therefore, tools that identify vulnerable patients preoperatively have clear clinical value. A growing body of evidence frames POD as a fundamentally neuroinflammatory disorder. Surgical trauma triggers a systemic inflammatory response that compromises blood-brain barrier (BBB) integrity, permits the infiltration of peripheral immune cells, and activates microglia, ultimately disrupting neuronal networks that subserve cognition [4,5]. This mechanistic understanding has fueled an intensive search for predictive and diagnostic biomarkers, spanning markers of Alzheimer pathology (phosphorylated tau, amyloid-β), neuronal injury (total tau, neurofilament light chain), astroglial activation (glial fibrillary acidic protein), systemic inflammation (interleukin-6), and BBB dysfunction (cerebrospinal fluid-to-serum albumin ratio) [5,6]. However, many of these methods require cerebrospinal fluid sampling or ultrasensitive immunoassays, which limits their utility at the bedside. Recently, serum amyloid A (SAA) has become a promising biomarker candidate. SAA is a major acute-phase protein (the circulating concentration can rise up to 1,000-fold during inflammation) that actively drives cytokine production, leukocyte recruitment, and endothelial dysfunction through receptors that include formyl peptide receptor 2 (FPR2), toll-like receptors 2 and 4 (TLR2/TLR4), scavenger receptors class B type 1 (SR-BI), and the receptor for advanced glycation end product (RAGE) [7]. Importantly, SAA is minimally expressed in the brain, so a circulating elevation most plausibly reflects a systemic pro-inflammatory state rather than central nervous system production, which is precisely the baseline vulnerability that may predispose the aging brain to delirium [7]. In this issue of the Korean Journal of Anesthesiology, Shin et al. [8] tested this hypothesis in a retrospective cohort of 731 patients aged ≥ 60 years undergoing hip surgery. POD occurred in 121 patients (16.6%) within five postoperative days. Preoperative SAA was markedly higher in the POD group (median 96.6 vs. 6.6 mg/L) and remained independently associated with POD after multivariate adjustment (odds ratio 1.004, 95% CI [1.002–1.007], P < 0.001) alongside age and preoperative albumin levels. Receiver operating characteristic curve analysis yielded an area under the curve of 0.812, with a threshold of 15.9 mg/L offering 81.9% sensitivity and 69.0% specificity. Notably, the same distinction was not apparent for postoperative SAA, which was markedly elevated in both groups and did not differ by delirium status (median 301.0 vs. 301.5 mg/L). Once surgery has occurred, the robust trauma-driven acute-phase response appears to overwhelm the more subtle inflammatory signal that distinguishes delirium-prone patients. Therefore, the preoperative measurement reflects the baseline inflammatory tone and carries the predictive weight. As an inexpensive and routinely available blood test with such an encouraging discriminatory performance, SAA could complement clinical risk stratification far more readily than cerebrospinal fluid-based markers. However, these findings should be interpreted with caution. The retrospective, single-center design of the study precludes the control of important confounders such as preoperative cognitive status, anticholinergic burden, and concurrent inflammatory markers (C-reactive protein, erythrocyte sedimentation rate) [9]. Restriction to hip surgery and a five-day window limits generalizability, and, as the authors acknowledge, elevated SAA is more likely to mark vulnerability than cause delirium. The bioactivity of SAA also depends on its lipidation state and high-density lipoprotein binding, which are variables that were not captured [7]. In conclusion, Shin et al. [8] provide encouraging evidence that preoperative SAA reflects neuroinflammatory vulnerability relevant to POD in elderly patients undergoing hip surgery. The next steps are clear: prospective, multicenter validation across diverse surgical populations; head-to-head comparisons with established inflammatory and neurodegenerative markers; and integration into multimodal, perhaps machine learning-assisted, risk models [5]. If these efforts succeed, SAA may gain a valuable place in the routine preoperative assessment of older surgical patients. Footnotes Funding None. Conflicts of Interest Jae Hoon Lee has been an editor for the Korean Journal of Anesthesiology since 2020. However, he was not involved in any process of review for this article, including peer reviewer selection, evaluation, or decision-making. There were no other potential conflicts of interest relevant to this article. References