Wearable Accelerometers for Predicting Perioperative Risk in Elderly Patients

Overview

This single-center prospective cohort study is designed to evaluate whether preoperative physical activity measured by a wearable accelerometer is associated with cardiovascular autonomic function and whether it can help predict post-induction hypotension in elderly patients undergoing elective non-cardiac surgery. Patients aged 65 years or older scheduled for general anesthesia with tracheal intubation will undergo preoperative physical activity monitoring using an ActiGraph wearable accelerometer. Activity-related parameters, including sedentary time, light-, moderate-, and vigorous-intensity activity, daily step count, and sleep duration, will be recorded. Cardiovascular autonomic function will be assessed perioperatively using noninvasive continuous hemodynamic monitoring, with baroreflex sensitivity-related indices derived from continuous blood pressure and beat-to-beat timing data. The primary outcome is post-induction hypotension. Secondary outcomes include early intraoperative hypotension, postoperative complications, and 30-day mortality. This study aims to clarify the relationship between habitual physical activity and autonomic regulation in older surgical patients and to provide evidence supporting future prehabilitation strategies.

Older adults undergoing surgery are at increased risk of perioperative hemodynamic instability and postoperative complications. Aging is associated with changes in cardiovascular autonomic regulation, including increased sympathetic activity and reduced parasympathetic tone. Perioperative autonomic dysfunction has been associated with adverse outcomes such as intraoperative hypotension, postoperative pain, postoperative nausea and vomiting, and impaired postoperative recovery. Baroreflex sensitivity is an important quantitative indicator of cardiovascular autonomic function and reflects the heart rate response to changes in blood pressure. Impaired baroreflex function in elderly patients may contribute to a higher risk of post-induction hypotension and other adverse perioperative outcomes. Physical activity has been shown to be associated with autonomic regulation, and exercise-based interventions may improve baroreflex sensitivity. However, the relationship between preoperative habitual physical activity and baseline autonomic function in elderly surgical patients has not been well defined. Wearable accelerometers provide an objective means of quantifying daily activity and may offer additional value beyond conventional self-reported functional assessments. This study therefore hypothesizes that preoperative physical activity level is associated with baseline cardiovascular autonomic function in elderly patients and that accelerometer-derived activity metrics may be useful in predicting post-induction hypotension and other perioperative outcomes. This study is a single-center, prospective observational cohort study. Eligible participants will be patients aged 65 years or older, classified as ASA physical status I to III, and scheduled for elective non-cardiac surgery under general anesthesia with tracheal intubation. Major exclusion criteria include severe vascular disease, secondary hypertension, limb tremor disorders such as Parkinson disease, inability to wear a wrist-based accelerometer, inability to measure upper-extremity blood pressure, arrhythmias unsuitable for baroreflex sensitivity measurement, and inability to cooperate with study procedures. Written informed consent will be obtained from all participants or their family members as required. Before surgery, participants will undergo collection of baseline clinical information, including comorbidities, current medications, and routine preoperative test results. Physical activity and frailty will also be assessed using the short-form International Physical Activity Questionnaire, the FRAIL scale, and the Fried frailty scale. After enrollment, participants will wear an ActiGraph wGT3X-BT accelerometer to continuously record daily physical activity and sleep-related parameters before surgery, including sedentary time, time spent in light-, moderate-, and vigorous-intensity activity, daily step count, and sleep duration. Perioperative hemodynamic monitoring will include preoperative noninvasive continuous arterial pressure monitoring using the LiDCO rapid V3 system. Under standardized resting conditions, continuous blood pressure and beat-to-beat cardiovascular data will be collected for baroreflex sensitivity analysis. Deep slow breathing testing will also be performed. Preoperative volume-related variables, including time since last fluid intake, intravenous fluid administration, and passive leg raising response assessed by stroke volume change, will be recorded. During surgery, routine intraoperative monitoring will include pulse oximetry, electrocardiography, and noninvasive blood pressure measurement every 3 minutes. Hemodynamic data at baseline, after induction, and at major intraoperative events will be recorded. Detailed information regarding anesthetic induction and maintenance, including sedative, analgesic, neuromuscular blocking, and vasoactive medications, will also be collected. Postoperative follow-up will be performed on postoperative days 1, 3, and 7 to assess complications according to the Clavien-Dindo classification, postoperative examinations, and quality of recovery using the QoR-15 questionnaire. A 30-day telephone follow-up will be conducted to assess postoperative complications, metabolic equivalent status, and other relevant events. The primary outcome is post-induction hypotension, defined as systolic blood pressure less than 90 mmHg, mean arterial pressure less than 65 mmHg, or a decrease of more than 30% from baseline within 20 minutes after induction of anesthesia or before surgical incision. Secondary outcomes include the incidence of early intraoperative hypotension, postoperative complications graded by the Clavien-Dindo classification, and all-cause mortality within 30 days after surgery. The planned sample size is 100 patients. Statistical analyses will evaluate associations between preoperative physical activity variables and post-induction hypotension, with adjustment for potential confounding factors such as induction regimen and baseline blood pressure using logistic regression models.