Prevalence of Pain in the Intensive Care Unit

Patients admitted to the intensive care unit (ICU), although often unable to communicate as a result of the severity of their disease process or the presence of impediments such as artificial airways or endotracheal tubes (ETTs), experience pain and discomfort that often goes unrecognized and untreated. Between 63% and 77% of patients report experiencing pain while in the ICU.

An observational study in 128 Italian ICUs of 661 postoperative patients found that 36.3% of patients did not receive any analgesia in the first 48 hours of their ICU stay. “Pain control” was the reason for the opioid administration in only 54.5% of instances in which an opioid was given. Fentanyl was the most commonly administered opioid. Although fentanyl’s rapidity of onset makes it perhaps desirable in an ICU setting, its use by prolonged infusion may be counterproductive in that its lipophilicity contributes to a prolonged offset of action when weaning or discontinuing this medication.

One study attempted to characterize the experience of a group of critically ill patients at high risk for hospital death. Between 55% and 75% of 100 cancer patients reported the experience of pain, discomfort, anxiety, sleep disturbance, or unsatisfied hunger or thirst.

One prospective cohort study of 400 medical ICU patients found that those patients for whom analgesics were prescribed had a higher concomitant incidence of hemodynamic monitoring, greater use of neuromuscular blocking agents, more mechanical ventilation days, and longer ICU and hospital lengths of stay. Consistent with these findings, patients who received analgesics also had higher trauma injury severity scores (TISS) and predicted mortality.

Barriers to Pain Assessment in the Intensive Care Unit

Given the potential compromise of patients’ physiologic stability and communication skills secondary to underlying disease processes, the ICU presents a unique environment for the assessment and treatment of pain. This presents unique challenges to clinicians that may not commonly be seen in other arenas of pain control. As mentioned, the presence of impediments to communication such as the presence of an endotracheal tube (ETT) or the severity of underlying critical illness may prevent typical ICU patients from communicating with the nursing or physician staff their level of discomfort ( Box 76.1 ). Indeed, critical care patients have a much higher risk of enduring untreated pain because they often cannot communicate secondary to altered mental status, mechanical ventilation, and sedation. A clinician’s observation and subjective rating of ICU patient pain often underestimates that patient’s pain, and whenever possible, ICU patients should rate their own pain.

A critically ill patient may be obtunded secondary to the underlying disease process or physiologically compromised by a process such as sepsis or shock. The question has been raised as to how much real pain the patient is experiencing, and this question is difficult to answer. Many patients may not remember these experiences, and even if these experiences can be recalled, there are limitations of assessing pain retrospectively. A chief challenge in assessing pain in the ICU comes when assessing patients in whom communication may be compromised. Increased vigilance by physicians and nursing staff, as well as the development and use of alternative means of pain assessment, are necessary when patients cannot verbalize pain.

One study examined the assessment and treatment of pain in two surgical ICUs in a university hospital setting. After the investigators found poor results in both of these areas, the following action plans of improvement were sequentially implemented: (1) education of the physician and nursing staff regarding the importance of pain and to measure pain with a modified visual analog scale (VAS) that is similar to a numeric rating scale ( Fig. 76.1 ), (2) assurance that these modified VAS scales were readily available at patients’ bedsides, (3) documentation of patients’ pain scores on daily ICU rounds, and (4) creation of an expectation in these ICUs that a pain score greater than 3 is a cause for intervention. Although 42% of nursing interval assessments were not measured on a standard 10-point scale at baseline, after the 5-week period of implementation of these strategies, pain assessments according to this scale increased to more than 70%. The study also found that patients whose pain scores were less than 3/10 increased from 59% before these interventions were begun to greater than 90% after they were completed.

Modified visual analog scale for pain.

Additional methods of evaluating pain in patients unable to communicate verbally have been developed from similar techniques in the pediatric arena. This would include the various behavioral pain assessment scales, in which pain scores are recorded by assessing a variety of patients’ nonverbal behaviors and sometimes including changes in vital signs. An investigation described earlier found that placing VAS cards at the patients’ bedsides permits some patients to point at a spot on the scale that corresponds to their pain score.

Moreover, pain in critically ill and postoperative pediatric patients is essential for their comfort and benefit, and the use of pain protocols in these regards improves patient comfort. More compliance and implantation of such protocols is very much needed in the pediatric critical care population. Another study attempted to compare three different pain assessment tools—the Critical-Care Pain Observation Tool (CPOT), the adult Nonverbal Pain Scale (NVPS), and the Faces, Legs, Activity, Cry, and Consolability scale (FLACC)—in nonverbal critically ill patients in a cardiac postanesthesia care unit. These scales were compared immediately before, 1 minute, and 20 minutes after two discomforting events: suctioning and repositioning. Both the CPOT and NVPS were found to be more reliable than the FLACC, which is predominantly used in a pediatric setting. After standardized implementation, the CPOT resulted in more a consistent assessment of pain in critically ill patients and fewer analgesic and sedative agents being administered.

One group attempted to validate a behavioral pain scale ( Fig. 76.2 ) for use in the ICU environment, noting that no pain scale for patients incapacitated by critical illness had yet been validated. Noting a significant difference between physiologic variables such as heart rate and mean arterial pressure, the observers were able to validate a behavioral pain scale at rest and during painful procedures. The behaviors evaluated in determining a score for pain were facial expression, upper limb movements, and compliance with mechanical ventilation. This scale was also felt to be feasible from a time standpoint, with an average assessment time of 4 minutes.

Behavioral pain scale.

Barriers to Pain Assessment in the Intensive Care Unit

Given the potential compromise of patients’ physiologic stability and communication skills secondary to underlying disease processes, the ICU presents a unique environment for the assessment and treatment of pain. This presents unique challenges to clinicians that may not commonly be seen in other arenas of pain control. As mentioned, the presence of impediments to communication such as the presence of an endotracheal tube (ETT) or the severity of underlying critical illness may prevent typical ICU patients from communicating with the nursing or physician staff their level of discomfort ( Box 76.1 ). Indeed, critical care patients have a much higher risk of enduring untreated pain because they often cannot communicate secondary to altered mental status, mechanical ventilation, and sedation. A clinician’s observation and subjective rating of ICU patient pain often underestimates that patient’s pain, and whenever possible, ICU patients should rate their own pain.

A critically ill patient may be obtunded secondary to the underlying disease process or physiologically compromised by a process such as sepsis or shock. The question has been raised as to how much real pain the patient is experiencing, and this question is difficult to answer. Many patients may not remember these experiences, and even if these experiences can be recalled, there are limitations of assessing pain retrospectively. A chief challenge in assessing pain in the ICU comes when assessing patients in whom communication may be compromised. Increased vigilance by physicians and nursing staff, as well as the development and use of alternative means of pain assessment, are necessary when patients cannot verbalize pain.

One study examined the assessment and treatment of pain in two surgical ICUs in a university hospital setting. After the investigators found poor results in both of these areas, the following action plans of improvement were sequentially implemented: (1) education of the physician and nursing staff regarding the importance of pain and to measure pain with a modified visual analog scale (VAS) that is similar to a numeric rating scale ( Fig. 76.1 ), (2) assurance that these modified VAS scales were readily available at patients’ bedsides, (3) documentation of patients’ pain scores on daily ICU rounds, and (4) creation of an expectation in these ICUs that a pain score greater than 3 is a cause for intervention. Although 42% of nursing interval assessments were not measured on a standard 10-point scale at baseline, after the 5-week period of implementation of these strategies, pain assessments according to this scale increased to more than 70%. The study also found that patients whose pain scores were less than 3/10 increased from 59% before these interventions were begun to greater than 90% after they were completed.

Modified visual analog scale for pain.

Additional methods of evaluating pain in patients unable to communicate verbally have been developed from similar techniques in the pediatric arena. This would include the various behavioral pain assessment scales, in which pain scores are recorded by assessing a variety of patients’ nonverbal behaviors and sometimes including changes in vital signs. An investigation described earlier found that placing VAS cards at the patients’ bedsides permits some patients to point at a spot on the scale that corresponds to their pain score.

Moreover, pain in critically ill and postoperative pediatric patients is essential for their comfort and benefit, and the use of pain protocols in these regards improves patient comfort. More compliance and implantation of such protocols is very much needed in the pediatric critical care population. Another study attempted to compare three different pain assessment tools—the Critical-Care Pain Observation Tool (CPOT), the adult Nonverbal Pain Scale (NVPS), and the Faces, Legs, Activity, Cry, and Consolability scale (FLACC)—in nonverbal critically ill patients in a cardiac postanesthesia care unit. These scales were compared immediately before, 1 minute, and 20 minutes after two discomforting events: suctioning and repositioning. Both the CPOT and NVPS were found to be more reliable than the FLACC, which is predominantly used in a pediatric setting. After standardized implementation, the CPOT resulted in more a consistent assessment of pain in critically ill patients and fewer analgesic and sedative agents being administered.

One group attempted to validate a behavioral pain scale ( Fig. 76.2 ) for use in the ICU environment, noting that no pain scale for patients incapacitated by critical illness had yet been validated. Noting a significant difference between physiologic variables such as heart rate and mean arterial pressure, the observers were able to validate a behavioral pain scale at rest and during painful procedures. The behaviors evaluated in determining a score for pain were facial expression, upper limb movements, and compliance with mechanical ventilation. This scale was also felt to be feasible from a time standpoint, with an average assessment time of 4 minutes.

Behavioral pain scale.

Barriers to Pain Treatment in the Intensive Care Unit

The difficulty of treating pain in those hospitalized in the ICU has been clearly documented, and there is a defined lack of high-level evidence for clinical decision making in these regards. Frequent need for rapid drug titration can contribute significantly to interpatient variability in response, which is likely more magnified in those who are critically ill. Moreover, there can be significant difficulty in distinguishing adverse drug events from other comorbid medical conditions in ICU patients.

Highlighting this clinical dilemma of treating pain in ICU patients, the SUPPORT investigators examined more than 4000 ICU patients over 2 years. In the data-gathering phase I portion, family members reported moderate to severe pain at least half the time for 50% of conscious patients who died in the hospital. Similar results were found in another study, in which a sample of 24 surgical patients from two different hospitals was interviewed after transfer from the ICU. Sixty-three percent of the patients rated their pain as being moderate to severe in intensity while in the ICU.

Another study analyzed the adequacy of pain control for 17 trauma patients during the initial aspect of their ICU admission. In contrast to 47% of patients rating their pain as severe, 95% of house staff and 81% of nurses reported the patients had received adequate pain control. This is underscored by the fact that 53% of house staff did not ask patients if pain control was satisfactory. The impact of insufficient analgesia was evidenced by the fact that 68% of patients reported that pain affected their ability to cough and 55% had trouble deep breathing. Presumed barriers to adequate pain control were felt to include a disparity in the perception of pain between patients and caregivers, patients refusing to request pain medicine despite the presence of moderate to severe pain, and physician and nurse concerns about patients’ adverse physiologic response to further or increased doses of analgesics. It was noted that 19% of patients reported a fear of addiction to opioids. Misconceptions regarding opioid pharmacology on the part of caregivers have been reported. In one report, some members of a nursing staff felt that the use of anxiolytics negated the need for larger doses of opioids.

A critically ill patient’s ability to communicate pain may be impaired by the severity of the injury or the presence of factors that impede communication, such as the presence of endotracheal or positive pressure ventilation. One study attempted to put forth guidelines regarding sedation and analgesia in the critically ill patient nearing death ( Box 76.2 ). In doing so, the panel recognized the difficulty of pain assessment in this setting as a result of several factors including communication problems particular to the ICU environment, the severity of critical illness and potential presence of multisystem organ failure, the possibility of a decreased level of consciousness resulting from illness and drugs, and difficulty in the interpretation and reporting of clinical signs. In the treatment of terminal disease, there are several possibilities for titrating sedation and analgesia: patient request, signs of respiratory distress, and physiologic signs such as tachycardia, hypotension, diaphoresis, facial grimacing, tearing, vocalization, or patient restlessness.

One investigation used a questionnaire to collect data on patients’ stressful experiences associated with being in an ICU and with mechanical ventilation. Of those who remembered the ETT, 68% were significantly bothered by not being able to speak (68%), pain associated with the ETT (56%), and anxiety regarding the ETT (59%). It is not surprising that this study found that those who did not remember the ETT or ICU were on average more severely ill and subject to a longer duration of mechanical ventilation than the group who remembered these experiences. It is quite possible, given the severity of their disease process, that the former group may have been more likely to be chemically paralyzed or heavily sedated.

Benefits of Assessing and Treating Pain in the Intensive Care Unit

ICU patients who are intubated may be at higher risk for pulmonary complications after surgery. Postoperative pain in these patients results in a pattern of shallow breathing with fewer sigh breaths. Complications of retention of secretion, atelectasis, hypoxemia, and pulmonary infections may be seen among this group. Epidural analgesia, intrapleural analgesia, intravenous patient-controlled analgesia (PCA), and intercostal nerve blockade (the duration of which is based on the duration of the local anesthetic used) are options in this situation. Clinical situations in which placement of thoracic epidural analgesia may be of significant benefit to the patient in the ICU include patients with post-thoracotomy pain; these catheters may be left in place to provide coverage of chest tube site pain until the chest tubes are discontinued. The placement of epidural catheters for pain control may aid trauma patients with rib fractures and also may help to facilitate weaning from mechanical ventilation.

PCA in the ICU is useful in an awake and alert patient by allowing the patient to receive on-demand dosing of analgesic agents, either via the epidural or intravenous route. Ballantyne and colleagues published a meta-analysis of randomized, controlled trials to examine postoperative analgesic therapies and their effects on pulmonary outcomes. Epidural opioids tended to decrease the incidence of atelectasis (respiratory rate [RR] = 0.53, 95% cardiac index [CI] = 0.33 to 0.85) and showed a less impressive tendency to decrease the incidence of pulmonary infections (RR = 0.53, 95% CI = 0.18 to 1.53) and overall pulmonary complications (RR = 0.51, 95% CI = 0.20 to 1.33) when compared with systemically administered opioids. Epidural local anesthetics, compared to systemically administered opioids, were found to decrease the incidence of pulmonary infections (RR = 0.36, 95% CI = 0.21 to 0.65) and overall pulmonary complications (RR = 0.58, 95% CI = 0.42 to 0.80). No clinically or statistically significant differences were found in other measures of pulmonary outcome, including forced expiratory volume in 1 second (FEV ₁ ), forced vital capacity (FVC), and peak expiratory flow rate (PEFR). This meta-analysis sheds light on the value of postoperative epidural analgesia in reducing postoperative pulmonary morbidity and questions the correlation of these spirometry values as predictors of postoperative pulmonary morbidity.

The benefit of controlling perioperative pain to attenuate the perioperative stress response has been debated. This response brings about an increase in sympathetic tone, increased secretion of catecholamines and catabolic hormones such as glucagon and cortisol, and decreased secretion of anabolic hormones such as insulin. It also increases antidiuretic hormone (ADH) and aldosterone secretion, electrolyte abnormalities, and protein catabolism. One theory suggests that opioids diminish the hyperglycemic and epinephrine responses because of efferent nerve blockade to the adrenal medulla yet have no effect on the cortisol response as a result of incomplete afferent nerve block.

One group attempted to study the quality of the dying process in the ICU by retrospectively interviewing family members of 38 ICU decedents. This investigation found that family members perceived that the patients experienced substantial physical symptoms during the last week of life, but they felt that their loved ones’ pain was under satisfactory control less than half the time. A review of the hospital records showed a near universal absence of data regarding pain assessment as well as records of how sedatives and analgesics were titrated to pain and symptom relief. Of the many items measured, the perception of how well the patient’s pain was under control correlated strongly with the quality-of-dying ratings as measured in the study. The authors concluded that end-of-life care efforts in the ICU include management of pain as well as support of dignity, respect, peace, and maximizing patient’s pain control.

Implementing a structured and scheduled approach to pain management in the ICU may also be of significant value, both to the patient and to the medical institution providing care. ICU patients whose pain regimens were titrated based on standardized sedation, analgesia, and delirium scores attained better analgesia, received fewer opiates, and, despite comparable sedation, endured a shorter duration of mechanical ventilation. Also, based on a wellness model and the World Health Organization’s multimodal analgesic ladder, pain outcomes after cardiac surgery were improved. Where “as needed” analgesics were previously implemented, scheduled dosing around the clock and prior to procedural pain provided effective pain relief in 95% of 120 patients over a 3-month period, for every ICU staff shift for the first 6 days after cardiac surgery postoperatively. Side effect profiles improved, and median length of stay after surgery decreased. Moreover, implementation of a structured ICU pain protocol at the University of Iowa medical ICU reduced the number of ventilator days required from 10.3 to 8.9, the cost per patient by $10,500, and a savings of nearly $2 million for the hospital over a 26-month period.