Abstract
Introduction
With the rapid development of intensive care medicine, more patients are recovering from critical conditions (Imanaka et al., 2010). Some intensive care unit (ICU) patients suffer from physical, cognitive, and mental dysfunctions that persist past the acute phase, known as post-intensive care syndrome (PICS). (Needham et al., 2012). PICS is accompanied by a decrease in patients’ quality of life (Briegel et al., 2013).
In Japan, PICS can also occur in patients who have undergone selective surgery for diseases such as esophageal cancer and acute myocardial infarction (Ohbe et al., 2021; Unoki et al., 2021)
One of the factors contributing to PICS is the stress associated with treatment, nursing care, the special environment of the ICU, and the stress of being under sedation with a ventilator (Hweidi, 2007; Needham et al., 2012; Pang & Suen, 2008; Puntillo et al., 2016; Yava et al., 2011).
To improve the long-term prognosis of patients, including their quality of life, the American Society of Intensive Care Medicine published a revised version of its “Clinical Practice Guidelines for the Management of Pain, Agitation, and Delirium in Adult Patients in the Intensive Care Unit” (Barr et al., 2013; Devlin et al., 2018). The guidelines include recommendations for types of sedatives, management of shallow sedation, and early rehabilitation.
It has been reported that following the guidelines can reduce the duration of ventilator placement (Kress et al., 2000), reduce ICU and hospital stay (Girard et al., 2008), improve patient mobility (Pohlman et al., 2010), and is beneficial in terms of healthcare costs (Dasta et al., 2010). The guidelines also mention nursing care; routine nursing care performed under sedation control, such as turning and endotracheal suctioning, can contribute to patient stress (Aslanidis et al., 2018; Gelinas et al., 2019; Grap et al., 2014; Siffleet et al., 2007).
Review of Literature
Turning and endotracheal suctioning are the most common aspects of nursing care performed on patients in the ICU. These are frequently performed as essential care for pressure ulcer prevention and respiratory drainage but can be stressful for the patient (Puntillo et al., 2014, 2018; Siffleet et al., 2007).
The assessment of stress and pain is based on patient complaints and self-reports. As a measurement of stress associated with nursing care, the behavioral pain scale and the critical-care pain observation tool are utilized (Gélinas, 2016). However, because most patients are sedated immediately after surgery, their responses to stimuli are suppressed (Shehabi et al., 2018); thus, it is difficult to communicate their stress verbally or through reactions (Devlin et al., 2018; Gélinas, 2016; Puntillo et al., 2017).
Other non-verbal and observable measures include serum biomarkers such as cortisol, epinephrine (Chlan et al., 2007), and salivary amylase (Grap et al., 2014). However, these are difficult to use to determine the stress associated with nursing care.
Other physiological indices such as vital signs (Arbour & Gélinas, 2010), level of arousal (Coleman et al., 2013; Gélinas et al., 2011), and autonomic nervous system indices (Bourgault et al., 2006; Chanques et al., 2017; Devlin et al., 2018) have also been reported as measures of stress in patients under sedation management. They are non-invasive and easy to use in unstable postoperative patients when assessing stress. They can also capture and measure trivial changes promptly, even in sedative patients who are slow to respond to stimuli (Chanques et al., 2017; Devlin et al., 2018).
This study aims to investigate patient stress caused by turning and endotracheal suctioning, routine nursing care performed in the ICU on sedated patients with esophageal cancer immediately after surgery, based on changes in physiological indices such as arousal level, autonomic nervous system indices, and vital signs.
Methods
Design
Observational Study Design
Research Question
We set the following research question prior to our investigation:
How is the stress associated with nursing care reflected in physiological indices in ICU patients with esophageal cancer whose responses to stimuli are suppressed by sedation?
Sample
Postoperative patients with esophageal cancer who were admitted to the ICU and underwent ventilator management from September to December 2019 were included in the study.
Inclusion Criteria
Postoperative patients with esophageal cancer admitted to the ICU who are under sedation management.
Exclusion Criteria
Patients with persistent preoperative arrhythmias, pacemakers, and anticholinergic medications were excluded because they would have precluded accurate data analysis when analyzing heart rate variability (HRV).
Definitions of Terms
This study defined stress as the physiological fluctuations in arousal level, autonomic nervous system indices, and vital signs resulting from the burden of endotracheal suctioning and turning. We also defined mean bispectral index (BIS) as the arousal level index, HRV as the autonomic nervous system index, mean blood pressure (MAP) and heart rate (HR) as the cardiovascular system index, and the central venous blood oxygen saturation (ScvO2) as the respiratory system index.
Collected Data Items
Patient age and sex were obtained from their health records. MAP, HR, and ScvO2 were obtained from the monitors worn (CSM-1702) (NIHON KOHDEN, Japan) for therapeutic management purposes. These data can be continuously extracted without new invasive procedures.
BIS was obtained from EEG data through its sensors (A-3200) (NIHON KOHDEN, Japan), and RR interval data was obtained from the wireless sensor RF-ECG2 and MemCalc/BonalyLight. All data were obtained via USB in Excel data files.
The data on the presence or absence of changes in the patient's facial expression and body movements during endotracheal suctioning and turning and the frequency of additional bolus sedation as a measure to reduce the burden of care were obtained by the researcher at the bedside and recorded in the field notes.
Calculation of Parasympathetic/Sympathetic Nervous System Indices
Frequency analysis was performed on the RR interval data using MemCalc software (GMS, Japan) to calculate the following HRV parameters: low frequency (LF; 0.04–0.15 Hz) and high frequency (HF; 0.15–0.40 Hz). HF was calculated as a parasympathetic index, and the ratio of LF to HF (LF/HF) as a sympathetic index (Task Force of The European Society of Cardiology and The North American Society of Pacing and Electrophysiology, 1996).
Data Collection Procedure
The investigators stayed at the patient's bedside without intervention from when the patients entered the ICU after surgery until they were extubated the next day. The investigators observed clinical nurses providing care to patients. Nursing care was classified into three patterns: turning and endotracheal suctioning, each performed separately and combined. The time required for these procedures was measured with a stopwatch, and care lists and patient behavioral indicators were recorded in field nodes. Instruments used to extract physiological index data were set to a predetermined date and time. For instruments whose date and time could not be changed, the date and time displayed on the screen were recorded and adjusted when analyzing the data.
In addition, since the HRV analysis devices needed to be newly attached to the patient, investigators confirmed that an environment was available for observation and obtained consent from the nurse in charge. Since the accurate analysis of HRV is not possible with HRV analysis, we excluded some supraventricular extrasystoles and artifacts from the data after confirming the electrocardiogram waveforms with the research collaborator, who is a physician.
Statistical Analysis
The results were averaged for each of the four points (1 min before, during, 1 min immediately after, and 5 min after). When averaging, we ensured the patient rested for at least 10 min before and after care. Multiple comparisons were made using the Wilcoxon signed rank test after performing the Friedman test for turning and endotracheal suctioning. The Shaffer method was used for correction. Due to the small sample size, we chose the Shaffer method, which allows for controlling Type II errors (Shaffer, 1995). The statistical significance level was set at 5%. Statistical analysis was performed using IBM SPSS Statistics (version 26).
Results
Sample Characteristics
Consent to participate in the study was obtained from 15 patients. One participant was excluded because it was difficult to accurately analyze the HRV owing to the persistent appearance of arrhythmia after surgery; thus, fourteen participants were analyzed. Ninety-nine scenes of nursing care were observed.
The participants were nine men and five women, with a mean age of 68 ± 8.9 years. The participants were controlled sedated from −4 to −5 on the Richmond Agitation-Sedation Scale (RASS). The sedatives used were propofol and dexmedetomidine.
During endotracheal suctioning and turning, the patients showed changes in behavioral indicators, such as facial expressions and movements of the upper and lower limbs, in 30 out of 99 scenes.
We observed six scenes in which a medical professional promptly administered additional bolus sedation to the patient during or after nursing care.
Single care: Endotracheal suctioning (Table 1)
Endotracheal suctioning was performed in 20 scenes. Of these, facial expression changes and coughing were observed in five scenes; no additional bolus sedation was administered in those five scenes. The average time required to perform endotracheal suctioning was 44.6 ± 5.4 s per session. No significant differences were found in all indices. However, LF/HF showed an increasing trend from before to during care in 14 of the 20 situations.
Effects of Endotracheal Suctioning at Different Time Points: Wilcoxon's Signed Rank Test With Shaffer Correction.
**
Single care: Turning (Table 2)
Turning was performed in 34 scenes. Of these, facial expression changes, upper and lower limb movements, and attempts to get up were observed during turning on two scenes; additional bolus sedation was performed on both of these scenes. Excluding those, 32 scenes were analyzed. The average time required for turning was 235.3 ± 24.3 s per change. Significant differences were observed in MAP, HR, and LF/HF, respectively. No significant differences were observed in BIS, ScvO2, and HF.
Effects of Turning at Different Time Points: Wilcoxon's Signed Rank Test with Shaffer Correction.
**
Combined care: Turning and endotracheal suctioning in succession (Table 3)
Turning and endotracheal suctioning were performed in 51 scenes. Of these, facial expression changes, coughing, and upper and lower limb movements were observed in 23 scenes; additional bolus sedation was administered in four of those scenes. Excluding those, 47 scenes were analyzed. The average time required to perform turning and endotracheal suctioning was 315.6 ± 20.0 s per session. Significant differences were observed in BIS, HR, ScvO2, and LF/HF, respectively. No significant differences were observed in MAP and HF.
Effects of Turning and Endotracheal Suctioning at Different Time Points: Wilcoxon's Signed Rank Test With Shaffer Correction.
**
Discussion
Type of Care and Magnitude of Stress
When turning and endotracheal suctioning were performed consecutively, significant changes were observed in all indicators except MAP, but when turning was performed alone, no change in arousal level was observed. Furthermore, when endotracheal suctioning was performed alone, no significant changes were observed in all indicators. These results suggest differences in the amount of stress applied to patients depending on the type and combination of care.
Endotracheal suctioning is perceived to be stressful care because an awake patient can recognize that the nurse is about to attempt endotracheal suctioning (Puntillo et al., 2014, 2018). Under sedation, the patient's inability to anticipate the onset of distress or the suppression of the cough reflex caused decreased stress. The respiratory index, ScvO2, was also unchanged, indicating that the stress caused by endotracheal suctioning was relatively small.
Turning is characterized by the dynamic effects of moving the patient's body. Patients in the acute postoperative period have an unstable cardiovascular status and are easily affected by changes in positioning (Brindle et al., 2013). Therefore, cardiovascular system indices such as MAP and HR may have changed.
This study showed that performing combined care may be more stressful than single care. This may be because the total care time was longer as it was 1 min longer than for turning alone. This longer care time may have induced greater stress on the sedated patient (Puntillo et al., 2014). To reduce patient stress, it may be necessary to consider performing them independently and shortening the time of care if not urgent and possible.
Contrary to expectations, no significant difference in MAP was observed when turning and endotracheal suctioning were performed consecutively. The result may be due to the order of turning and endotracheal suctioning. In this study, the combined care had a mix of patterns of endotracheal suctioning after turning and vice versa. The effect of endotracheal suctioning after turning may have mitigated the effect of hemodynamic changes on blood pressure, making it more difficult to detect significant changes.
Type of Index and Their Usefulness
Autonomic nervous system indices showed an increasing trend in 14 of 20 situations during endotracheal suctioning. Significant changes were observed in other care, suggesting that they are more sensitive than other indices (Broucqsault et al., 2016; Devlin et al., 2018; Chanques et al., 2018).
Conversely, the arousal level index showed significant changes only when combined care was performed. This is consistent with increased arousal levels in nursing care, as shown in previous studies (Gelinas et al., 2011; Li et al., 2009), suggesting that the index had low sensitivity but high specificity for major stress (Arbour et al., 2015; Gelinas, 2016; Gelinas et al., 2011). Therefore, the care that causes fluctuations in arousal level has a high risk of causing significant stress to the patient and is recommended to be more carefully applied.
Vital signs may be used as an index to infer what specific stresses are occurring in the patient by evaluating a composite of several indicators. For example, when combined care was performed, BIS and HR began to increase during care, and ScvO2 changed after the end of care. It has been reported that shallower sedation increases oxygen consumption and increases the tendency to arousal (Terao et al., 2003). This could indicate that the care load stimulated the sympathetic nervous system, increasing arousal level and HR and ultimately increasing oxygen consumption. The cardiovascular system index may reflect stress quickly, while the respiratory system index may lag behind changes in cardiovascular indicators and reflect stress. Paying attention to which vital signs are changing and at what time may help identify the specific stresses occurring in the patient.
Measures to Reduce Stress
Typically, when a patient is determined to be experiencing stress, a drug intervention is used in the ICU to reduce the patient's stress (Devlin et al., 2018). The criterion for nurses to implement measures to reduce patient stress from care is often whether there is a change in the patient's behavioral indicators (Gélinas et al., 2019). This study also administered a bolus dose of sedatives as a stress reduction measure. However, only six doses were administered in the 30 cases in which patients showed changes in facial expression or limb movements. The BIS of the six patients who received the additional dose was more than 10 higher than the remaining 24 patients who did not receive the additional dose. This means that patients may have been more responsive to the stress imposed by nursing care, and the behavioral changes in patients may have been more evident. The 24 cases in which no additional doses were given may have been less responsive, and the changes may have been less noticeable to the nurses who were focused on nursing care.
Previous studies have reported changes in vital signs such as HR, MAP, and SvO2 due to turning (Arbour and Gélinas, 2010; Li et al., 2009; Payen et al., 2001; Shively, 1988; Winslow et al., 1990). These studies were conducted on awake patients with higher values of change in vital signs compared to the present study. Patients under sedation management are generally reported to have suppressed responses to stimuli (Gelinas, 2016; Li et al., 2009; Payen et al., 2001). Since the participants in this study were under sedation management of RASS −5 to −4, the sedative drugs might suppress the fluctuations in vital signs. While this is a beneficial phenomenon in postoperative management (Devlin et al., 2018), it may be difficult to detect stress, including pain, associated with nursing care (Aissaoui et al., 2005; Robleda et al., 2016). Moreover, changes in arousal levels and vital signs continued during and after nursing care. This is similar to previous studies (Arbour & Gélinas, 2010; Erden et al., 2018; Li et al., 2009; Winslow et al., 1990). Nursing care is generally addressed to improve the patient's condition, so the damage accompanying nursing care is not often recognized. Physiological indicators may be useful in catching such potentially existing stresses. If we can focus our attention not only on behavioral indicators but also on physiological indicators and use them effectively, we may be able to better identify the occurrence of stress added to the patient and provide mitigating measures.
Strengths and Limitations
This study was conducted at a single institution. Moreover, it was a small number of participants and did not consider individual backgrounds such as the stage of esophageal cancer, drugs used, medical history, pain, respiratory control, noise, and illumination, which are considered to be related to the results. Furthermore, since this was an observational, field-based study, the techniques and order of turning and endotracheal suctioning were inconsistent. Therefore, the generalization of the findings is limited. In the future, it is necessary to increase the number of cases, paying attention to the types of diseases, individual backgrounds, and uniformity of techniques. Conversely, this study provides important insights into how nursing care provided to patients under sedation management may affect physiological indicators. This is a fundamental finding for future consideration of the management of pain, agitation, and delirium in the ICU.
Implications for Practice
This study suggests that routine nursing care for patients under sedation immediately after surgery may be stressful for patients. The distress and stress in patients under sedation control are controlled to some degree but not eliminated. Proactive recognition by nurses using these indicators of potentially occurring stress and providing more considerate care will improve the quality of life of postoperative patients.
Conclusions
This study examined changes in arousal level, autonomic nervous system indices, and vital signs associated with turning and endotracheal suctioning, which are essential care provided by nurses to sedated patients in the postoperative ICU.
The results revealed that different types and combinations of nursing care might cause different patient stress. Moreover, the autonomic nervous system indices may be more likely to react to stresses in a variety of nursing care, while arousal levels may be more likely to react to burdensome stresses. If the characteristics of these physiological indicators can be effectively utilized with behavioral indicators, it may be possible to accurately identify stress in patients with suppressed responses to stimuli and provide mitigating measures.
Footnotes
Acknowledgments
Declaration of Conflicting Interests
Funding
Ethical Consideration
ORCID iDs
