Results
A total of 984 patients who had elective open or laparoscopic abdominal surgery were enrolled in the underlying trials, 570 from FACTOR and 414 from EXPLANE. Our analysis was restricted to 673 patients among the 914 who had ViSi Mobile monitoring (Figure 1). Baseline medical, demographic, anesthetic, and surgical data are presented in Table 1, divided by time-weighted average pain scores.
Figure 1.
Flowchart of study population. EXPLANE, Transversus Abdominis Plane Block with Liposomal Bupivacaine versus Continuous Epidural Analgesia for Major Abdominal Surgery: the EXPLANE clinical trial; FACTOR, Effect of Intravenous Acetaminophen on Postoperative Hypoxemia After Abdominal Surgery: the FACTOR Randomized Clinical Trial.
The median [interquartile range] monitoring duration (after removal of gaps) was 32 [23 to 40] h during the first 48 postoperative hours. Overall, patients spent a median [interquartile range] of 7% [3 to 13%] of the total monitoring time sitting or standing, corresponding to 1.7 [0.7 to 3.1] h/day.
In patients with time-weighted average pain scores 3 or less, mobilization time was 8% [4 to 15%], corresponding to 1.9 [1.0 to 3.6] h/day. In patients with time-weighted average pain scores between 3 and 6, mobilization time was 7% [3 to 13%], corresponding to 1.7 [0.7 to 3.1] h/day. In patients with time-weighted average pain scores 6 or greater, mobilization time was 5% [2 to 11%], corresponding to 1.2 [0.5 to 2.6] h/day (Figure 2). Each unit increase in time-weighted average pain score was associated with an adjusted median decrease of 0.12 (97.5% CI, 0.02 to 0.24) h/day (P = 0.009; Table 2).
Figure 2.
Mobilization time based on time-weighted pain scores/opioid use. (Left) Mobilization time in hours per day, based on time-weighted average pain score during the initial 48 postoperative hours. (Right) Mobilization based on opioid use in milligrams of intravenous morphine equivalents during the initial 48 postoperative hours. Boxes represent interquartile ranges; middle bars, medians; diamonds, averages; whiskers extend to the most extreme value within 1.5 times the interquartile range below the first or above the third quartile. More extreme values (outliers) are represented by circles. iv, intravenous.
The opioid consumption ranged from 8 [0 to 50] mg of morphine equivalents in patients with low pain scores to 67 [34 to 171] mg of morphine equivalents in patients with the highest pain scores (Table 1). The main opioids used and their morphine equivalents are shown in Table A2. There was no significant association between postoperative opioid consumption and mobilization time, with an estimated adjusted median change of −0.04 (97.5% CI, −0.12 to 0.08) h/day for a twofold increase in morphine equivalent opioid consumption (P = 0.508; Figure 2; Table 2).
There was no significant association between time-weighted average pain score or opioid consumption and postoperative mobilization among patients more than 65 yr old (Table 2). We did not find significant associations between time-weighted average pain score and standing position; opioid consumption was inversely associated with standing, although not by a clinically meaningful amount (estimated adjusted median change of −0.01 [97.5% CI, 0.01 to −0.004] hours per day with each twofold increase in opioid consumption). Daytime (from 7 AM to 10 PM) mobilization was inversely associated with both time-weighted average pain score and opioid consumption. The estimated adjusted median change was −0.40 (97.5% CI, −0.72 to −0.08) hours per daytime day for each unit increase of time-weighted average pain score and −0.32 (97.5% CI, −0.56 to −0.08) hours per daytime day for each doubling of opioid consumption. Mobilization during the first 24 postoperative hours was not associated with time-weighted average pain score or with opioid consumption.
We found a significant interaction between pain score and surgical approach (P = 0.033; Table 2). For patients who had open procedures, each unit increase in pain score was associated with 0.17 (97.5% CI, 0.06 to 0.28; P <0.001) fewer hours of mobilization per day; this association was no longer significant for patients who had laparoscopic surgery, where each unit increase in pain score was associated with 0.05 more hours per day (97.5% CI, −0.15 to 0.25; P = 0.628). We did not find an interaction between morphine use and the surgical approach (P = 0.586). After additional adjustment for surgical approach in our primary outcome analysis, the results were similar to our primary analysis, where a doubling of morphine use was associated with 0.04 (97.5% CI, −0.14 to 0.05) fewer hours of mobilization per day (Table 2).
The composite of postoperative complications was observed in 17 patients, 9 of whom had pulmonary complications; 3 had myocardial injury; 1 had a stroke/transitional intravascular accident, and 4 had venous thromboembolism (Table 3). Considering quartiles of mobilization time, the incidence of the composite outcome was 6.0% (10 of 168 patients) among patients who spent 0 to 0.7 h in mobilization per day, 4.2% (7 of 168) among patients who spent 0.7 to 1.6 h in mobilization per day, and none among the remaining 337 patients in the highest two quartiles, who had more than 1.6 h/day. There was thus a significant association between mobilization time and postoperative complications with an estimated adjusted odds ratio of 0.34 (95% CI, 0.16 to 0.72) for each hour increase of mobilization per day, adjusting for age, sex, race, and surgery duration; P = 0.005. Preoperative American Society of Anesthesiologists Physical Status, intensive care unit admission, length of hospital stay, surgical approach, and surgery duration were also summarized by mobilization time.
Anesthesiology. 2022;136(1):115-126. © 2022 American Society of Anesthesiologists | Lippincott Williams & Wilkins
