Association Between the Preoperative Triglyceride-Glucose Index and Intraoperative Blood Pressure Variability in...

Background: Intraoperative blood pressure variability (BPV) is associated with adverse perioperative outcomes, although the prognostic meaning of variability itself, once blood pressure level is accounted for, remains uncertain. Preoperative metabolic indicators have not been examined in relation to intraoperative BPV. The triglyceride-glucose (TyG) index, a surrogate marker of insulin resistance, is associated with arterial stiffness in community populations. We examined whether the preoperative TyG index is associated with intraoperative BPV in surgical patients, and whether an estimated pulse wave velocity (ePWV) surrogate statistically accounts for part of this association. Methods: This single-center retrospective cohort study included 7081 adults who underwent elective major abdominal surgery under general anesthesia for at least 120 min at Beijing Tsinghua Changgung Hospital between January 2016 and December 2022. The primary outcome was the coefficient of variation of intraoperative mean arterial pressure (CV-MAP). ePWV was calculated from age and preoperative mean arterial pressure (MAP) and is therefore a surrogate rather than a direct measure of arterial stiffness. Multivariable linear regression assessed the TyG-CV-MAP association; models excluding and including age are reported as co-primary analyses. Exploratory mediation analysis using the Baron-Kenny framework with 5000 bootstrap iterations estimated the proportion of the association statistically accounted for by ePWV. Subgroup analyses tested age as an effect modifier, and a sensitivity model additionally adjusted for antihypertensive and antidiabetic medication use. Results: After adjustment for sex, body mass index, comorbidities, surgical duration, American Society of Anesthesiologists physical status, and surgery type, each unit increase in the TyG index was associated with a 0.0128-unit increase in CV-MAP (95% confidence interval [CI], 0.0097 to 0.0158; p < 0.001); with age additionally included, the estimate was 0.0086 (95% CI, 0.0056 to 0.0117; p < 0.001). In absolute terms, each unit increase in TyG corresponded to a 1.074 mmHg higher standard deviation of intraoperative MAP (MAP-SD; 95% CI, 0.842 to 1.305), against a cohort mean MAP-SD of 12.0 mmHg. ePWV statistically accounted for approximately one quarter to one third of the total effect (indirect effect, 0.0038; bootstrap 95% CI, 0.0030 to 0.0046). The association was present in patients younger than 65 years (β = 0.0132, p < 0.001), with no evidence of an association in those aged 65 years or older (β = 0.0008, 95% CI, -0.0054 to 0.0071; p = 0.79; interaction β = -0.0124, 95% CI, -0.0187 to -0.0061; p < 0.001). Sensitivity analyses supported the robustness of these findings, including a model adjusted for antihypertensive and antidiabetic medication use (β = 0.0128, 95% CI, 0.0098 to 0.0159; p < 0.001). Conclusions: The preoperative TyG index was independently associated with greater intraoperative BPV in major abdominal surgery, with the association present in patients younger than 65 years but not in older patients. An ePWV-based surrogate statistically accounted for approximately one quarter to one third of this association; because ePWV is derived from age and blood pressure rather than measured directly, this proportion may reflect the structure of the surrogate rather than a causal arterial-stiffness pathway. The clinical significance remains uncertain given the modest incremental predictive contribution and the absence of clinically anchored endpoints; prospective validation with directly measured pulse wave velocity is required.