Ask the Experts

Q What is the current evidence on pain and sedation assessment in nonresponsive patients in the intensive care unit?A Denise Li, RN, MS, CCRN, and Kathleen Puntillo, RN, DNSc reply:The management of pain and anxiety/agitation is an emerging standard of care for critically ill patients.1–3 Nonresponsive critically ill patients cannot provide self-reports of pain. Therefore, objective pain indicators must be used to achieve the timely management of the adverse effects of pain and to evaluate the effectiveness of interventions. Pain in nonverbal patients can be perceived as a suffering associated with a procedure, bodily injury, or disease that is characterized by physical and/or emotional discomfort, which gives rise to a set of distinctive behaviors perceived by caregivers as indicative of that discomfort.4 Analgesia aims to reduce pain perception and thus decrease behavioral or autonomic responses to noxious stimuli.Various definitions of sedation exist.5–7 In the intensive care unit (ICU), sedation often signifies one of the therapeutic effects of pharmacological agents that are used to alleviate anxiety and agitation and to induce a calm tranquil state.8 Hence, sedation intends to produce a state of sleep that includes a lack of awareness and a lack of recall.Little research has been reported on valid pain indicators in nonresponsive patients (ie, patients who are deeply sedated or comatose). Nevertheless, certain caregiver observations may provide insights about pain in sedated patients. Notably, pain behaviors correlate with acute pain intensity.9,10 Patients with procedural pain also displayed a greater number of pain behaviors than did patients without procedural pain.11 Five objective pain measures are available for the assessment of pain in sedated and/or critically ill adult patients12–16; however, only the Behavioral Pain Scale14 has been evaluated in sedated ICU patients receiving mechanical ventilation. The Behavioral Pain Scale consists of 3 behavioral domains (ie, facial expression, movements of upper limbs, and compliance with ventilation). These domains have differential responses depending on whether stimuli are nociceptive or non-nociceptive.14 In this study, mean arterial pressure and heart rate did not show clinically important changes during nociceptive procedures.Nurses using a pain assessment algorithm noted pain-related behaviors in ICU surgical patients.15 The most frequently observed pain-related behaviors, noted by greater than 20% of 14 nurses, were no movement, grimacing, wincing, vocalization, and restlessness. Slow, decreased movement, tenseness, attention-seeking movements, wrinkled forehead, and splinting of pain site were reported less frequently (12%–17%). Rigidity, being drawn around the mouth, and teariness/crying constituted less than 10% of the total behaviors reported. Increased heart rate (30%) and blood pressure (26%) were also noted. Other physiological indicators such as tachypnea, decreased heart rate, decreased blood pressure, perspiration, and pallor were reported in less than 15% of all observations.A retrospective chart review was done to examine indicators of pain noted by nurses and physicians.17 In 183 pain occurrences in 52 patients during the first 3 days of intubation, the pain indicators most frequently recorded by nurses were body movements (59%), cardiovascular responses (15%), compliance with ventilator (10%), and respiratory responses such as tachypnea and oxygen desaturation (10%). In physicians’ notes, patients’ reactions to physical examination (eg, withdrawal, localization of pain) and body movements (eg, agitation) were most often recorded. Only a few noted compliance with the ventilator as a pain indicator. In contrast to prior findings,15 facial pain expressions and neuromuscular signs (eg, rigidity) were seldom documented in this study. Whether these pain responses were not as evident in this particular Canadian sample or not considered by clinicians as valid pain indicators is not known.A recent report documented various pain behaviors observed in adult ICU patients undergoing 6 procedures (ie, femoral sheath removal, central venous catheter placement, tracheal suctioning, wound care, wound drain removal, and turning).11 Five behaviors were commonly associated with procedural pain, namely grimacing, eyes closed, rigidity, wincing, and verbal complaints. The validity of these behaviors as pain indicators was supported by significant differences between patients with procedural pain and without procedural pain who demonstrated them (P<.001). A moderate and significant correlation was found between patients’ self-reported pain intensity and the number of facial movements, verbal responses, body movements, and total behaviors (P < .001). Compared with patients without procedural pain, the probability of patients with procedural pain exhibiting these behaviors was nearly 3 times as high for increased facial responses, 4 times as high for increased body movement responses, and 10 times as high for increased verbal responses.Together, these findings suggest that behavioral responses are valid pain indicators in ICU patients. Considerations of factors that may affect the validity of these pain behaviors for use in sedated patients warrant further investigation. However, although facial pain expressions (eg, grimace, frown) were evident in studies of sedated patients receiving mechanical ventilation14 and intubated surgical ICU patients,15 certain factors may obscure the ability to assess pain-related behaviors. Older patients’ aging appearance and the effects of sedatives may obscure facial expressions.18 Moreover, physical restraints and other physical devices that are commonly used to protect ICU patients19–21 are likely to preclude the patients from freely moving their extremities. According to the author of the Behavioral Pain Scale (J. Payen, oral communication, October 14, 2003), none of the study patients was physically restrained. Pain protective movement (eg, slow, decreased, or no movement) as observed in oriented patients,15 may not be evident in deeply sedated patients. As for physiological indicators of pain, there is inconclusive support that such indicators (eg, vital sign changes, perspiration, pupil dilatation) have validity, mainly because they lack specificity as pain indicators.Sedation is primarily used in ICUs to treat anxiety or agitation. Anxiety is the psychophysiological response to the anticipation of real or imagined danger, and agitation is excitement accompanied by motor restlessness.3 Two types of subjective sedation scales are currently available. Scales that assess alertness or arousability (eg, Observer’s Assessment of Alertness/Sedation)22–24 are useful to assess opioid-induced sedation. Another type of subjective scale is the type of scale designed to evaluate a patient’s need for sedation therapy. Such scales include the Ramsay Sedation Scale,25 the Motor-Activity Assessment Scale,26 the Sedation-Agitation Scale,27 and the Richmond Agitation-Sedation Scale.28 These scales have descriptors that extend over the entire range of the level of consciousness continuum and include 3 domains (ie, agitation, normal cognition, decreased level of consciousness). To date, 4 sedation scales are known to have preliminary evidence of validity and reliability when used in adult ICU patients.28–30Typical indicators of agitation found on the scales include a gradual increase in patients’ level of restlessness or nonpurposeful movement and a decrease in cooperativeness (eg, pulls at endotracheal tube or catheters, climbs out of bed, is uncooperative). Patient-ventilator dyssynchrony is used by the Richmond Agitation-Sedation Scale as an indicator of agitation.32 Restlessness and decreased cooperation were shown to have significant correlations (P<.001) with agitation as indicated on the Motor-Activity Assessment Scale.26 Patients deemed agitated on the basis of the Sedation-Agitation Scale were more intolerant to mechanical ventilation than were patients not deemed agitated.27 Sedation indicators on these scales include a gradual decrease in arousability (eg, lack of eye contact) as well as behavioral responses to voice, tactile, or noxious stimuli.Assessing pain and sedation in nonresponsive patients is challenging. A major challenge is the confounding effect of sedation on objective indicators of pain. Clinicians might infer that adequate sedation means different patient states: promotion of amnesia, sleep/rest, patient safety, ventilator synchrony, and hemodynamic stability.32,33 Hence, an ideal measure that can adequately address the complexity and individualize the nature of the goals of pain and sedation therapy remains elusive.34 Furthermore, the behavioral responses to pain and anxiety/agitation (eg, restlessness, ventilator dyssynchrony, and movement) have many similarities. Tolerance to mechanical ventilation has been suggested to have validity in both an ICU pain scale14 and a sedation scale.31 Additional research is needed to establish the validity, sensitivity, and specificity of these pain indicators in sedated patients. In the meantime, in circumstances where patients are nonresponsive to external stimuli, clinicians should integrate other information such as actual or potential risks of pain (eg, extensiveness of injury, invasive therapies, intubation) and risks of pain-related functional impairment into their pain assessment in nonresponsive, sedated patients.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.