Journal: medRxiv
Article Title: Wearables Anticipate Postoperative Complications: A Prospective Cohort Study
doi: 10.64898/2026.06.02.26354556
Figure Lengend Snippet: Mean trajectories (solid lines) and 95% confidence intervals (shaded areas) are shown for patients with major complications (blue) and minor or no complications (gray) across postoperative days, with values expressed as change from preoperative baseline. (A) Resting heart rate, (B) maximum pain, (C) sleep temperature, and (D) daily step count are displayed. Vertical dashed lines indicate the day of surgery (postoperative day 0). Across all metrics, patients with major complications exhibited slower and more prolonged deviations from baseline compared with patients with minor or no complications. For each postoperative day, group-level means and standard errors of the mean (SEM) were computed across all participants with available preoperative baseline values and valid daily measurements for the corresponding metric. To improve visualization of longitudinal trends, daily mean trajectories and SEM were smoothed within each complication group using a 3-point centered rolling mean (window = 3 days; minimum observations = 1).
Article Snippet: In the first 10 postoperative days, peak absolute deviation from baseline was higher in the major-complication group for Garmin resting heart rate (22.2 ± 7.1 vs 14.5 ± 7.1 bpm; P=0.005, Mann–Whitney U), Oura resting heart rate (26.2 ± 12.5 vs 17.7 ± 10.1 bpm; P=0.046, Mann–Whitney U), Oura sleep temperature deviation (1.53 ± 0.81 vs 1.00 ± 0.54 °C; P=0.043, Mann–Whitney U), and Oura readiness score (43.4 ± 15.6 vs 29.7 ± 12.4; P=0.034, Welch t-test).
Techniques: