End-of-life predictions in advanced cancer: nurses’ perspectives and the possible role of contact-free vital sign monitoring—a prospective exploratory observational study

Introduction

Cancer is a major cause of death worldwide (1). Medical staff treating patients with advanced cancer should provide appropriate support for the maintenance of a proper life and also a “good death” for these individuals (2,3). Life expectancy prediction is essential for medical staff to provide appropriate care and treatment at the optimal time and for the caregivers/families to prepare for the coming event (4). Many previous studies have focused on predicting last days. In association with impending death, these studies identified clear objective changes in vital signs such as blood pressure reduction, heart rate increase, saturation of peripheral oxygen (SpO₂) reduction, and neurobehavioral changes such as decreased responses to external stimuli and nonreactive pupils (5-11). Being able to predict the last two weeks of life and inform patients and their families of this may lead to a better, more fulfilling end, but there is little research into predicting this period.

Experienced nurses working in palliative care units accurately detected patients’ worsening condition as a sign of impending death (12,13). Nursing records and actions taken are based on their perception of the worsening of the patient’s daily condition. Thus, investigating nursing records could reveal what action and when nurses perceive as signs of a terminally ill cancer patient’s “last weeks”. Furthermore, by evaluating changes in vital signs alongside these records, we may detect small but significant deterioration in vital signs that correlates with nursing actions.

Vital signs and physical symptoms are fundamental indicators used to evaluate the condition and physical performance of patients with advanced cancer; however, wearable or cable monitors become a burden in deteriorating patients at the end of life, such that most palliative care units do not utilize them (14,15). We developed a contact-free non-restraining bed sensor system (BSS) for continued vital sign monitoring (16) and successfully analyzed respiratory and behavioral features in patients receiving palliative care and opioids for cancer related symptoms (17,18). This BSS is capable of monitoring not only respiration but also small weight fluctuations from heartbeat to large weight changes such as body movements on the bed and bed leaving of the patients. Seow et al. [2011] showed that the Palliative Performance Scale declined sharply 4–6 weeks before death (19). Hui et al. reported that a decrease in consciousness level emerged a week ahead of death in 97% of patients with advanced cancer and patients who were completely bed-bound (7,8). Using the BSS, we recently demonstrated significantly smaller body movements on the bed in advanced cancer patients with performance status 4 than in patients with performance status 3 (18). Accordingly, we wondered whether nurses possibly sense the reduced physical performance on the bed during the last weeks which can be captured by the BSS.

This study aimed to clarify the following two points: (I) to investigate the actions of nurses in the palliative care ward of our center which may predict the last two weeks of life; (II) to assess whether the most common nurses’ actions related to their perception of the worsening of the patient’s general condition during the period can be manifested in the vital signs captured by the noncontact device under the bed. Specifically, we hypothesized that the perception of the patient’s deterioration would coincide with the reduced body movement activity on the bed detected by the device. We present this article in accordance with the STROBE reporting checklist (available at https://apm.amegroups.com/article/view/10.21037/apm-25-95/rc).

Methods

Study population

This prospective observational exploratory study was performed at the Cancer Center Hospital in South Kanto area, Japan. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This prospective observational study was approved by the Ethics Committee of Chiba Cancer Center [IRB No. R03-004]. Written informed consent was obtained from each patient admitted to the palliative care unit after explaining the risks and purpose of the present study. The study was registered at the UMIN Clinical Trial Registry (UMIN000046214, Principal researcher: Yoshihiko Sakashita, Registration Date: November 29, 2021). Patient recruitment was performed between January and September 2022. Among the cancer patients admitted to the palliative care unit, we recruited adult patients with advanced cancer who had a palliative prognostic score of ≥4. The exclusion criteria were a diagnosis of head and neck cancer and permanent or transient tracheostomy, since respiratory patterns, the key observation during sleep, was suspected to alter due to changes in airway morphology.

Evaluation of nurses’ action changes

During the study period, 23 nurses with at least three years of experience in acute care wards were assigned to a 28-bed palliative care unit. Only on-the-job training (OJT) by experienced nurses was required to work in this department; no special training was necessary. Based on previous reports and our institutional experience (8,13), we predetermined nine medical chart descriptions that strongly reflect nurses’ action changes that could occur when they predict impending death (Table 1). The predictive descriptions were searched for and identified by the primary investigator (D.S.) from the medical chart from admission to discharge. The day of the first description in the medical chart was determined. We considered the prediction of 3–14 days (last weeks) before death provided the families and caregivers as well as the patient sufficient time to prepare for the death that this was therefore clinically meaningful for achieving a good death, and preferable to a prediction at 0–2 days (last days) before death. Accordingly, among the nine predictive descriptions in the medical chart, we identified the predictive description with the highest event rate between 3–14 days before death. The behavioral changes were recorded as an event of “the day” when the behavior occurred between 6 a.m. that day and 6 a.m. the following day. Nurses were unaware of the purpose of the study. We ensured that no intention was made explicit regarding data collection.

Monitoring with BSS

Continuous data sampling from BSS system (bed sensor vital sign monitoring system, BSS Minebea-Mitsumi, Nagano, Japan (Figure 1) was performed from admission to discharge. The precise functionality of the BSS system has been explained in our previous reports (16-18,20). In brief, four load cell sensors were placed under the wheels of the medical bed and continuous signals were sent to the data logger for data processing and analysis. The four load cells independently measured the total weight of sensors that detected small movements, such as centroid shift due to heartbeats, respiration, and slight body movements to total bed leaves (i.e., when the patient completely left the bed). The magnitude of the abrupt weight shift on the sensors was calculated as the acceleration (g/s²) and named the activity index (ACI). Other vital signs included respiratory and pulse rates. After calculating the respiratory rate and pulse rate, instability of the respiratory rate and pulse rate was assessed by relative changes in respiratory rate and pulse rate to those of the preceding breath or pulse. Body weight changes, measured in grams, were continuously monitored, and the number of bed leaves was monitored based on an abrupt change in body weight to 0 kg during the monitoring period. All signals and variables were processed at a 10 Hz sampling rate within the data logger and stored in a personal computer connected to the data logger for subsequent analyses. The data from the daily BSS were divided into two periods: (I) daytime 6:00 a.m.–10:00 p.m., and (II) nighttime 10:00 p.m.–6:00 a.m. The median of each period was calculated for each BSS variable, except for the ACI, which used averaged hourly data. The uneven time allocation is intentional, separating wakefulness from sleep to make the difference clear.

Figure 1 BSS used in this research is shown (original photo). Sensors are placed under the four wheels and signals are collected in the logger under the bed (not visible in this photo) from which data are sent to the computer installed with a program to analyze specific waves. The monitor, shown in this photo, was not placed in the patient’s room but a personal computer was connected to the logger for later analysis. Real time monitoring of respiratory rate, bed occupancy status, and patient movement in bed is possible. BSS, bed sensor system.

Aims of the study

First, among the nine predetermined nurses’ predictive descriptions for impending death on the medical chart, we aimed to determine the one with the highest event rate within 3–14 days before death (last day). The event day on which the predictive description with the highest event rate was written on the medical chart was also determined. The second aim was to test the hypothesis that body movement activity on the bed measured by BSS (the primary variable) decreases on the event day relative to the three days prior to the event day.

Sample size calculation

In a study of 51 patients admitted to a palliative care ward, the ACI from 10:00 p.m. to 6:00 a.m. was 33±31 g/s² (17). We expected a 17 g (50%) decrease in the nocturnal ACI on the event day relative to the ACI on the three days prior to the event day. For α=0.05, 1-β=0.8, and standard deviation (SD) =31 g/s², the required sample size was calculated to be 29 patients. Considering the possible dropout cases, the number of participants enrolled was set at 40.

Statistical analysis

Data were analyzed using SigmaStat version 14.0 (Systat Software Inc., Point Richmond, CA, USA). Continuous variables are presented as medians [25%, 75% interquartile ranges (IQR)], and categorical variables are presented as frequencies and percentages. The days before death were treated as continuous variables. No treatment for missing values was planned. Comparison of the continuous variables between the event day and the day three days prior to the event day was performed using the Wilcoxon signed-rank test. The event rates were compared using Fisher’s exact test. When the event occurred immediately after 1–2 days of admission, variables at admission were used for comparison. P values of <0.05 were considered to indicate statistical significance, and all p values were two-sided.

Results

Forty elderly Japanese cancer patients receiving end-of-life care during the study period were enrolled. Five patients were excluded from the analysis because the BSS data were not successfully collected due to technical failure in 3 patients, and discharge from the palliative care unit to their home in 2 patients. Thus, the medical charts and BSS data from 35 patients were analyzed. Table 2 presents the patients’ background characteristics. The primary cancer site was the gastrointestinal system in 66% of the patients with metastasis to various vital organs. The patients were in the terminal stage of cancer; 66% had a Karnofsky Performance Scale of ≤40 and 91% had a palliative prognosis score of >5.6, suggesting a life expectancy of less than 30 days. Their vital signs measured by nurses on admission were stable, and oxygen was administered to 34% of the patients.

Palliative care nurses’ actions before death during admission

Nine predetermined actions by palliative care nurses were identified on medical charts with various timings and frequencies. As shown in Figure 2, palpation of the radial artery (57%) and cessation of oral medication (42%), possibly due to the detection of worsening conditions, were the top two most likely to be documented during the last weeks before the last days than other predetermined nurses’ actions. On the other hand, the start of time series records (69%) and reporting medical conditions to the night shift supervisor (54%) were the top two key actions during the last days for preparing for the impending death. Accordingly, we considered palpitation of the radial artery as the most common nurse action associated with the deterioration of the patient during the last weeks before death. Vital signs measured during the daytime by nurses on the day of radial artery palpation event did not differ from those measured 3 days prior to the event, while SpO₂ significantly decreased by 1% on the day of radial artery palpation (Table 3).

Figure 2 Time and frequency of the nurses’ nine predetermined events first documented on the medical chart. Values represent the number of events recorded on the medical chart and the percentages and median (interquartile range). *, P<0.05; **, P<0.01 vs. “palpation of the radial artery” event frequency, respectively. IQR, interquartile range.

Changes of BSS variables before and after nurses’ palpation of the radial artery

Table 4 presents the vital signs measured by the BSS variables on the day of the radial artery palpation event and those 3 days prior to the event. BSS monitoring revealed that the patients rarely left their beds during the day or at night, but the ACI reflecting the magnitude of body movements on the bed was greater during the daytime than at night, indicating the presence of a circadian rhythm. In disagreement with our original hypothesis, the daytime ACI did not decrease on the day of the radial artery palpation event relative to the daytime ACI in the three days before the event (P=0.57), although the frequency of daytime bed leaving significantly decreased (P=0.003). In contrast, the nighttime ACI slightly, but significantly, increased on the day of radial artery palpation (P=0.046). Both pulse rate and respiratory rate continuously measured by the BSS were slower than those measured once during the day by nurses and did not differ between the day of radial artery palpation and three days before the event. Notably, the respiratory rate varied by >10% on both the day of radial artery palpation and three days before the event, and this respiratory instability significantly increased during the night of radial artery palpation (P=0.02).

Discussion

This prospective observational study conducted in a palliative ward of a cancer hospital in Japan found that “palpitation of the radial artery” is the most common nurse’s action in association with the deterioration of the patient during the last weeks before death (3–14 days before death). Continuous non-contact monitoring revealed small but significant increases in respiratory rate instability, nocturnal ACI, and daytime hours spent in bed associated with the nurses’ decision to perform radial artery palpation. The SpO₂ level decreased slightly before and after palpation of the radial artery, but no statistically significant changes were observed in the other vital signs.

This study revealed that palpation of the radial artery is the first action taken when death is suspected, compared to other nursing interventions. Discontinuing medication, the second-highest documentation within 2 weeks before death found in this study, could be part of the care tailored to the rapidly deteriorating condition of terminal cancer patients. This result agrees with Chen’s analysis that dysphagia was a significant predictor of survival for <2 weeks (13). Inquiring about the family or supporters’ awareness of the patient’s death was an important form of support for the families of cancer patients who were nearing the end of life. Additionally, close coordination with team members regarding the patient’s worsening condition could lead to a shared understanding of death within days or hours among all involved (21).

In this study, nurses performed radial artery palpation, although they were aware that the blood pressure and pulse rate, usually monitored once a day in the morning, were within the normal range. This indicates that palpitation of the radial artery is a common act when nurses suspect or sense some deterioration of the physical status towards death. Ikari et al. identified a palpable radial artery as one of the characteristics of unexpected survival for >3 days (22). Pulselessness of the radial artery was one of reasons why the palliative care physicians expected death within 24 hours (23). In non-cancer patients, the prediction model “pulselessness of the radial artery OR respiration of mandibular movement OR Shock Index >1.0” predicted death within 7 days with an accuracy of 83.9% (24). This study confirmed these results and further suggests that the emergence of nurses’ behavior of confirming a palpable radial artery would be a convincing sign of their recognition that the patient would die within few weeks.

Experienced nurses working in palliative care units accurately detected the patient’s worsening condition as a sign of impending death (12,13). Accuracy improves with proximity to death indicating less accuracy for predicting weeks and months before death. The usefulness of surprise questions for nurses in predicting long-term and short-term survival has been reported in palliative care settings (22,25). Although they inform patients’ families in preparation for coming death, the actual signs that nurses sense from the patients have not been well clarified (23). Changes in intermittently measured vital signs, such as blood pressure and pulse rate, have been observed during the last 48 h of life (8). Thus, the results of this study further demonstrated the contribution of nurses in accurately predicting “last weeks before death, which is an essential yet challenging task for all medical staff.

The usefulness of continuous non-contact vital sign monitoring before death for predicting impending death has recently been reported (26), such as, increases in pulse rate, respiratory rate, and apnea. Although some vital sign trends do not match our findings, the importance of continuous monitoring in detecting such changes is also supported by this study. We found minor but significant changes in the ACI, nocturnal respiratory instability, and percentage of daytime bed use at 3–14 days before death. An increase in respiratory instability may be in accordance with apneas detected by Tanaka et al. (26). When death is approaching, patients become wholly bedridden and spend most of their time in bed during the daytime. Consequently, nighttime sleep becomes shallow, and the nocturnal ACI increases slightly but significantly. Before emerging apparent deteriorating vital signs, nurses would likely sense the combination of these small changes in physical activity and predict a worsening of the general condition approaching death within 2 weeks.

The present study had several strengths. The non-contact monitor used in this study allowed us to continuously gain vital signs while the physical conditions deteriorated. To the best of our knowledge, this study is the first to detect an increase in respiratory rate instability when nurses sensed the change in patients’ condition and decided to palpate the radial artery in terminally ill cancer patients. The change was not detectable with respiratory rate monitoring by the nurses’ visual confirmation or rate monitoring from the BSS. Regarding respiratory rate instability, a novel index reported by our group calculating relative changes in respiratory wave intervals, has been reported to increase with increasing opioid dosing (17) or worsening performance status (18).

Our study had several limitations. First, its small sample size, inclusion of only Japanese patients, and single-center design could limit the generalizability of our findings. The sample size was calculated based on the hypothesis that the ACI would decrease. However, these results did not support our hypothesis. A single-center design limits the use of drugs and nurses’ experience. A multi-institutional study with a larger number of participants is required to detect and rule out the effects of drugs and environmental changes. Second, although the accuracy of respiratory parameters of the BSS used in this study has been validated (16,27), other parameters, such as ACI and percent time in bed, have not been validated with gold-standard methods, such as actigraphy. Nevertheless, in our data obtained from post-surgery (28) or oncology patients with different performance statuses (18), the ACI have been shown to accurately represent what is clinically observed. Therefore, we believe that our data are reliable. Finally, the accuracy of the measurement of an increase in nocturnal ACI, respiratory rate instability, and percentage of daytime hours spent in bed as predictors of deteriorating conditions in cancer patients must be confirmed in future studies.

Conclusions

In conclusion, the act of radial artery palpation for the first time in patients with advanced cancer is likely to signify approaching death within 3–14 days. No clinically significant difference was found in the intermittent vital signs measured by the nurse between the time of the action and three days before the event. However, increased respiratory instability, nocturnal in-bed ACI, and time spent in bed during the daytime, as detected using a non-restrained and non-contact vital measurement device, increased significantly. Continuous and precise measurements of vital signs may, in the future, assist medical experts in the diagnosis of approaching death and therefore improve the care for families of dying patients.

Acknowledgments

We express our sincere gratitude to the nursing staff of the Palliative Care Unit, Chiba Cancer Center, for their assistance in conducting this study.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://apm.amegroups.com/article/view/10.21037/apm-25-95/rc

Data Sharing Statement: Available at https://apm.amegroups.com/article/view/10.21037/apm-25-95/dss

Peer Review File: Available at https://apm.amegroups.com/article/view/10.21037/apm-25-95/prf

Funding: This study was supported by a grant-in-aid. JSPS KAKENHI (21K09007 to D.S.).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://apm.amegroups.com/article/view/10.21037/apm-25-95/coif). D.S. reports that the study was supported by JSPS KAKENHI, Grant Number 21K09007 (Chief Investigator: D.S.), paid to Chiba Cancer Center. S.I., as an ex-head of the Department of Anesthesiology at Chiba University, received scholarship donations from Minebea Mitsumi, Inc. He was an inventor of the vital sign monitoring system used in this study. The scholarship donation from Minebea Mitsumi Inc. is planned for Article Processing Charge payment for this article. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Chiba Cancer Center (IRB No. R03-004). Written informed consent was obtained from each patient admitted to the palliative care unit after explaining the risks and purpose of the present study.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 2021;71:209-49. [Crossref] [PubMed]
2. Miyashita M, Sanjo M, Morita T, et al. Good death in cancer care: a nationwide quantitative study. Ann Oncol 2007;18:1090-7. [Crossref] [PubMed]
3. Leung KK, Tsai JS, Cheng SY, et al. Can a good death and quality of life be achieved for patients with terminal cancer in a palliative care unit? J Palliat Med 2010;13:1433-8. [Crossref] [PubMed]
4. Hiratsuka Y, Hamano J, Mori M, et al. Prognostication in advanced cancer: foreseeing from global insights and foretelling in the Asian context. Jpn J Clin Oncol 2025;55:991-9. [Crossref] [PubMed]
5. Bruera S, Chisholm G, Dos Santos R, et al. Variations in vital signs in the last days of life in patients with advanced cancer. J Pain Symptom Manage 2014;48:510-7. [Crossref] [PubMed]
6. Mori M, Morita T, Bruera E, et al. Prognostication of the Last Days of Life. Cancer Res Treat 2022;54:631-43. [Crossref] [PubMed]
7. Hui D, dos Santos R, Chisholm G, et al. Clinical signs of impending death in cancer patients. Oncologist 2014;19:681-7. [Crossref] [PubMed]
8. Hui D, Dos Santos R, Chisholm G, et al. Bedside clinical signs associated with impending death in patients with advanced cancer: preliminary findings of a prospective, longitudinal cohort study. Cancer 2015;121:960-7. [Crossref] [PubMed]
9. Hwang IC, Ahn HY, Park SM, et al. Clinical changes in terminally ill cancer patients and death within 48 h: when should we refer patients to a separate room? Support Care Cancer 2013;21:835-40. [Crossref] [PubMed]
10. Mori M, Yamaguchi T, Maeda I, et al. Diagnostic models for impending death in terminally ill cancer patients: A multicenter cohort study. Cancer Med 2021;10:7988-95. [Crossref] [PubMed]
11. Mori M, Yamaguchi T, Maeda I, et al. Prediction of Next-Day Survival in Imminently Dying Cancer Patients: A Multicenter Cohort Study. J Palliat Med 2025;28:345-50. [Crossref] [PubMed]
12. White N, Reid F, Vickerstaff V, et al. Specialist palliative medicine physicians and nurses accuracy at predicting imminent death (within 72 hours): a short report. BMJ Support Palliat Care 2020;10:209-12. [Crossref] [PubMed]
13. Chen PY, Huang CH, Peng JK, et al. Prediction Accuracy Between Terminally Ill Patients' Survival Length and the Estimations Made From Different Medical Staff, a Prospective Cohort Study. Am J Hosp Palliat Care 2023;40:440-6. [Crossref] [PubMed]
14. Chester JG, Rudolph JL. Vital signs in older patients: age-related changes. J Am Med Dir Assoc 2011;12:337-43. [Crossref] [PubMed]
15. Pavic M, Klaas V, Theile G, et al. Mobile Health Technologies for Continuous Monitoring of Cancer Patients in Palliative Care Aiming to Predict Health Status Deterioration: A Feasibility Study. J Palliat Med 2020;23:678-85. [Crossref] [PubMed]
16. Isono S, Nozaki-Taguchi N, Hasegawa M, et al. Contact-free unconstraint respiratory measurements with load cells under the bed in awake healthy volunteers: breath-by-breath comparison with pneumotachography. J Appl Physiol (1985) 2019;126:1432-41. [Crossref] [PubMed]
17. Hasegawa M, Nozaki-Taguchi N, Shono K, et al. Effects of opioids on respiration assessed by a contact-free unconstraint respiratory monitor with load cells under the bed in patients with advanced cancer. J Appl Physiol (1985) 2021;130:1743-53. [Crossref] [PubMed]
18. Nozaki-Taguchi N, Takai H, Shono K, et al. Continuous monitoring of activity and vital signs with load cells under the bed legs in advanced cancer patients: a prospective exploratory observational study-can it represent performance status? Ann Palliat Med 2023;12:757-66. [Crossref] [PubMed]
19. Seow H, Barbera L, Sutradhar R, et al. Trajectory of performance status and symptom scores for patients with cancer during the last six months of life. J Clin Oncol 2011;29:1151-8. [Crossref] [PubMed]
20. Yoshimura A, Nozaki-Taguchi N, Suganuma D, et al. Respiratory Effects of Benzodiazepine in Patients with Advanced-Stage Cancer Receiving Opioid Analgesics: A Prospective Observational Study. Palliat Med Rep 2024;5:399-407. [Crossref] [PubMed]
21. Sekse RJT, Hunskår I, Ellingsen S. The nurse's role in palliative care: A qualitative meta-synthesis. J Clin Nurs 2018;27:e21-38. [Crossref] [PubMed]
22. Ikari T, Hiratsuka Y, Yamaguchi T, et al. "3-Day Surprise Question" to predict prognosis of advanced cancer patients with impending death: Multicenter prospective observational study. Cancer Med 2021;10:1018-26. [Crossref] [PubMed]
23. Bruera S, Chisholm G, Dos Santos R, et al. Frequency and factors associated with unexpected death in an acute palliative care unit: expect the unexpected. J Pain Symptom Manage 2015;49:822-7. [Crossref] [PubMed]
24. Hosoi T, Ozone S, Hamano J, et al. Prediction Models for Impending Death Using Physical Signs and Vital Signs in Noncancer Patients: A Prospective Longitudinal Observational Study. Palliat Med Rep 2021;2:287-95. [Crossref] [PubMed]
25. Suh SY, Yoon SJ, Lin CP, et al. Are Surprise Questions and Probabilistic Questions by Nurses Useful in Home Palliative Care? A Prospective Study. Am J Hosp Palliat Care 2024;41:431-41. [Crossref] [PubMed]
26. Tanaka H, Fukui S, Maeda I, et al. The change over time of vital signs with consideration for opioid use in the last 2 weeks of life among cancer patients in a palliative care unit: Continuous measurement of vital signs using a non-wearable monitor. Cancer Med 2021;10:8799-807. [Crossref] [PubMed]
27. Inada A, Inaba S, Matsumura Y, et al. Contact-free assessments of respiratory rate and volume with load cells under the bed legs in ventilated patients: a prospective exploratory observational study. J Appl Physiol (1985) 2023;134:1341-8. [Crossref] [PubMed]
28. Son K, Tarao K, Hateruma Y, et al. Perioperative continuous body weight measurements with load cells under the bed legs in patients undergoing abdominal surgery. Eur J Anaesthesiol Intensive Care 2023;2:e0028. [Crossref] [PubMed]