| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
DISTINGUISHED RESEARCH LECTURE |
Abstract
Assessment and management of patients’ pain across practice settings have recently received the increased attention of providers, patients, patients’ families, and regulatory agencies. Scientific advances in understanding pain mechanisms, multidimensional methods of pain assessment, and analgesic pharmacology have aided in the improvement of pain management practices. However, pain assessment and management for critical care patients, especially those with communication barriers, continue to present challenges to clinicians and researchers. The state of nursing science of pain in critically ill patients, including development and testing of pain assessment methods and clinical trials of pharmacological interventions, is described. Special emphasis is placed on results from the Thunder Project II, a major multisite investigation of procedural pain.
Along the Road to Knowledge Development
In the Wizard of Oz,1 Dorothy followed the Yellow Brick Road to get to the Wizard’s Emerald City. This road was laced with both roadblocks and breakthrough discoveries. The Yellow Brick Road contained many twists and turns, and Dorothy met memorable characters along the way.
The path to knowledge about pain assessment and management in critically ill patients also contains many twists and turns. However, those clinicians and researchers who have been traveling with me on this ever-widening path of discovery recognize how far we have come in a relatively short time. Basic science has provided us with a clearer and more detailed understanding of pain mechanisms, and clinical science has provided us important knowledge about pain measurement and management. Travelers on the path know that the pace of discoveries is accelerating, just as we are seeing an acceleration of knowledge related to critical care.
Critical Care Units and the Science of Pain: Separate Paths
Our modern critical care units developed from the interests and needs of clinicians to have particular areas in which the care of patients with specific and difficult problems was concentrated.2 For example, in the 1940s, trauma units were developed to manage war casualties. In the 1950s and 1960s, postoperative recovery rooms, respiratory care units, and units for patients who had had cardiac surgery were developed to provide specialized care to specific populations of patients. These intensive care units (ICUs) were furnished with the best technologies and the best trained staff available. But, as noted by Christopher Bryan-Brown,2 during this evolutionary stage, pain management was never considered a primary mission for critical care. In fact, Bryan-Brown and some of his medical colleagues drew organized attention to the issue of pain in the critically ill in the 1986 publication of the book Acute Pain Management.3 The authors characterized unrelieved pain as the major psychological and physiological stressor that it is and summarized the extant knowledge about pharmacological methods that could be considered for critically ill patients. The authors suggested several interventions to diminish the presence and psychological impact of pain. I had recently begun my doctoral studies on pain in the critically ill, and this text had a major impact on my own work. Before this time, the effects of various analgesic interventions on pulmonary function had been examined in a number of medical studies,4–7 but the investigators omitted the measurement of pain. The results of another series of studies8–14 reinforced that pain was a major stressor for ICU patients, but few studies had pain as the primary focus. As late as the mid-1980s, research that directly and specifically targeted and measured pain in critically ill patients was almost nonexistent.
Along the Road to Discovery
To investigate the impact of pain on patients who had been in ICUs, my colleagues and I followed the advise of Glinda, the good witch from the Wizard of Oz, who noted, "It’s always best to start at the beginning."1 Thus, in the late 1980s, we conducted a descriptive study15 in which we explored the following questions during interviews of 24 patients after their stay in an ICU:
A total of 71% of the patients surveyed remembered having pain, and 63% rated it as moderate to severe in intensity. They described pain from postoperative incisions and from procedures such as endotracheal suctioning, removal of chest tubes, and intravenous injections of potassium. One man described his incisional pain during endotracheal suctioning as follows16(p4): "I would be stretched out like a board, gagging and coughing, and the pain was intense. I felt taut and tense; the coughing is what made it so painful; it makes your whole body go rigid into the cough, and that was excruciating."
Patients reported numerous problems in communicating their pain to nurses, and perhaps as a result, patients were very conservatively medicated. That is, the mean amount of morphine administered to a sub-sample of 9 patients after cardiac surgery was 14 mg/d for the first 3 days in the ICU. These findings helped focus the problem of pain in the critically ill and provided direction to other studies in the early 1990s, including my own research for my doctoral dissertation.
The research for the dissertation included a systematic assessment of the postoperative17 and procedural18 pain of patients undergoing cardiac or abdominal surgery. Instruments used to assess pain, previously validated in non-ICU populations of patients, were used to quantify multiple dimensions of patients’ pain. In a group of 74 patients, the intensity of surgical pain was moderate, and patients who had abdominal surgery experienced more pain than did patients who had cardiac surgery.17 Interestingly, the pain did not diminish during the first 3 days after surgery. The lack of pain relief over time may have been due, in part, to the small amount of analgesics that the patients received overall and the decrease in amount each day (ie, from a mean of 14.4 mg of morphine-equivalent opioids on day 1 to 6.4 mg on day 3). That pain can indeed increase morbidity was inferred by an additional finding that patients with higher pain scores had significantly more atelectasis than did patients with lower pain scores (P < .05).
Two other major findings came from the dissertation research. First, the painfulness of 2 commonly performed procedures was quantified.18 On a numeric rating scale of 0 to 10, pain associated with endotracheal suctioning was rated as moderate (mean, 4.9). The pain associated with chest tube removal (mean, 6.6) almost reached an intensity recognized as severe19 (ie, 7–10). Again, the lack of preemptive administration of analgesics almost certainly confounded the patients’ pain responses. Most patients had received no analgesics in the hour before the endotracheal suctioning or chest tube removal, and when opioids were administered, morphine-equivalent doses were small (mean, <3 mg).
Second, the research indicated that many ICU patients could use several different pain instruments to characterize the multidimensionality of their pain experiences. In so doing, patients provided extensive information about the intensity, location, and quality of their pain and made qualitative sensory and emotive distinctions between the characteristics of surgical versus procedural pain. Many of the patients were able to use these instruments despite being intubated. Thus, ICU practitioners now had a scientific basis for doing as extensive a pain assessment as possible, by using similar instruments, as a prerequisite to good pain management.
The early 1990s saw the publication of 2 books that provided guidance to clinicians who cared for acutely or critically ill patients in pain. One resulted from the vision of Kathleen Dracup, editor for a series of publications on critical care issues and a valued member of my dissertation committee. Through Dr Dracup’s leadership, some of my colleagues and I contributed the book Pain in the Critically Ill: Assessment and Management20 to her series. The second book, Acute Pain Management: Operative or Medical Procedures and Trauma,21 published by the former Agency for Health Care Policy and Research, provided clinical practice guidelines. These books offered state-of-the-science data useful to clinicians in the care of acutely or critically ill patients in pain.
Forks in the Road
In the mid-1990s, my research teams and I began to focus on 3 areas of inquiry:
Nonverbal Measures of Pain
To examine the accuracy of critical care nurses’ pain assessments, my research colleagues and I at the University of California, San Francisco (UCSF), developed the pain assessment and intervention notation (P.A.I.N.) tool. This tool was an algorithm that contained a checklist of behavioral and physiological indicators of pain. In subsequent sections of the algorithm, prompts were provided to assist the study nurses in deciding how to treat pain. Participating nurses were asked to use this tool when they did their first assessment of a patient after abdominal or thoracic surgery and for 4 subsequent hourly assessments. The behavioral or physiological indications of pain identified by the patients’ nurses at least 20% of the time included the following: no movement; grimacing, frowning, wincing; increased heart rate; increased blood pressure; and vocalization.22 The degree of accuracy of the nurses’ pain assessments was indicated by the concordance between nurses and their patients concerning the patients’ pain. That is, the pain intensity scores determined by the 2 groups (patients and nurses) did not differ significantly. Nurses’ pain ratings were consistently lower, however, than patients’ pain ratings across several pain assessment times. Furthermore, the doses of opioid analgesics that nurses chose to administer to their patients correlated more often with the nurses’ ratings of a patient’s pain than with the patient’s own self-reports. Although this study indicated the effectiveness of a focused assessment of patients’ behavioral and physiological signs of pain in increasing the accuracy of pain assessment, the study also revealed the complexity of clinical decision making for pain interventions.
Clinical Judgment and Knowledge of ICU Clinicians
Nuances of clinical decision making were explored in a qualitative analysis of nurses’ judgments and actions while they were using the P.A.I.N. algorithm.23 Nurse participants were asked to "think aloud" and to audiotape these responses while using the algorithm. Using interpretive phenomenology, we analyzed 31 tape recordings of 14 nurses caring for 41 patients. From this analysis, we learned the following:
These findings indicate that pain assessment is a cognitive and intuitive process in which a patient’s private pain experience is often played out within a social context, in this instance, an ICU environment.
The P.A.I.N. algorithm was a step in the right direction in that it provided a framework for organizing data on assessment and management of pain. In a detailed evaluation of the algorithm,24 participating nurses noted that the tool provided a consistent, systematic approach to quantifying their clinical judgment about pain, patients’ responses to pain, nurses’ choices of analgesics, and patients’ responses to those analgesics. The nurses who participated in the study thought that the tool was too long and cumbersome for everyday clinical practice and recommended that it be shortened. However, many nurses thought that their experience with using the tool would have a lasting effect on their practice. Almost all thought that such a tool would be especially helpful to critical care nurses early in the nurses’ practice.
These findings provided direction for future inquiries. For example, we continue to ask what type of pain assessment tool would be specific enough so that clinicians would feel more confident that what they are recognizing in their patients is indeed pain? Does such a tool capture the multidimensionality of pain by allowing an assessment of the intensity and location of pain and the distress associated with the pain, the qualities of pain, and behavioral and physiological responses to pain? Could use of this tool be part of an intervention to improve advocacy for patients, collaboration and negotiation with ICU team members, and, ultimately, patients’ pain and clinical outcomes? These questions have yet to be explored.
One natural extension of our previous work was a more detailed examination of the pain and sedation practices of critical care physicians and nurses, because the 2 practices are so often intricately linked. I was delighted to have the opportunity to work on such a study in the ICUs at UCSF with clinical nurse specialist Hildy Schell-Chaple and physician Neal Cohen, who was medical director of the ICUs at that time. These ICUs constitute my critical care practice and primary research base. We wanted to know what signs and symptoms were used by house staff and critical care nurses to determine an ICU patient’s need for analgesics or sedatives while the patient was receiving or not receiving neuromuscular blocking agents. We also explored the pharmacological knowledge of the 49 nurses and 34 physicians from 4 ICUs in our medical center who responded to the survey. Although the report from this survey is still in progress, noteworthy findings include the following:
It is clear from these findings that some ICU practitioners have specific knowledge deficits that warrant attention. The publication of practice guidelines27 on use of sedatives and analgesics in adults by a task force of the Society of Critical Care Medicine in 2002 was an important contribution to efforts to provide needed education about use of these medications.
Effectiveness of Interventions for Pain Associated With Removal of Chest Tubes
Having previously described the painfulness of chest tube removal, my colleagues and I decided to intervene. We knew, from a national survey28 of critical care nurses conducted by Marguerite Kinney, Karin Kirchhoff, and me, that national practices associated with removal of chest tubes varied widely. The 553 respondents to the survey noted that an analgesic prescription was seldom routinely available, and many nurses wished that pain management during chest tube removal was better.
The first interventional study,29 published in 1996, was a randomized clinical trial of injection of either bupivacaine or a placebo (isotonic sodium chloride solution) through pleural chest tubes before removal of the tubes in a sample of 41 patients after cardiac surgery. Testing of bupivacaine was based on the hypothesis that blockade of thoracic intercostal nerves by the regional anesthetic agent would interfere with pain transmission. However, no significant differences were found in pain intensity, distress, sensation, or affect scores between the group receiving bupivacaine and the group receiving the placebo. Yet, as often happens along a road to discovery, we had a serendipitous finding. Pain associated with removal of chest tubes was significantly lower in patients who had received intramuscular ketorolac, a nonsteroidal anti-inflammatory agent, a mean of 3 hours before the procedure than it was in patients who had not received ketorolac.
This finding helped shape our most recent double-blind, randomized clinical trial (K. Puntillo, J. Ley, unpublished data, 2003). In this trial, we gave patients intravenous injections of morphine (4 mg) or ketorolac (30 mg) and timed the chest tube removal to each drug’s peak effect. Pain intensity and distress levels at the time of removal of the chest tube did not differ significantly according to type of analgesic administered. However, the mean procedural pain intensity and distress scores of 3.2 (SD, 3.0) and 2.98 (SD, 3.18), respectively, in the sample as a whole indicated mild pain (K. Puntillo, J. Ley, unpublished data, 2003). Such a reduced level of pain intensity had not been reported in other studies of pain associated with the removal of chest tubes.18,29,30–35 We think that our newest findings offer clinicians latitude in their decision to use either an appropriately dosed and timed opioid or nonopioid non-steroidal anti-inflammatory agent in the preemptive management of pain associated with the removal of chest tubes.
The Crossing of Paths: Thunder Bolts and Lightning Rods
By the mid-1990s, the state of the science of pain in the critically ill had shown significant progress. First, the prevalence, significance, and complexity of pain assessment and management had been established. Second, valid, reliable, and useful assessment tools that provided accurate measurements of the multiple dimensions of pain had been tested in critically ill patients. Nevertheless, further research was necessary to establish better methods of assessing pain in nonverbal patients. At the same time, the evidence was mounting that pain associated with commonly performed procedures in acute and critical care settings was an important clinical concern. Thus, AACN, in its vision and concern for optimal care for patients, chose to support a study of procedural pain, soon referred to as the Thunder Project II. The research team of the Thunder Project II acted as a lightning rod to focus the creative research energy of the project and ground the study in scientific principles and clinical practice. As a team, we conceptualized, designed, and implemented the project and analyzed and disseminated its results.
Thunder Project II was a comprehensive, descriptive study of pain perceptions and responses of patients to 6 common procedures: turning, removal of wound drains, tracheal suctioning, removal of femoral catheters, insertion of central venous catheters, and nonburn wound dressing change. Trained volunteer nurse research coordinators and research associates at each of 169 sites across the United States and in Canada, the United Kingdom, and Australia obtained data from 6201 patients aged 4 to 97 years. This project resulted in the collection of extensive information about patients’ pain intensity, location, and quality; behavioral and physiological responses; procedural distress; and analgesics administered. Thus far, the research team has 3 publications36–38 and 4 abstracts39–42 in print. Six publications are in review or in progress, and an additional 3 or more are planned. Members of the research team have presented findings at 1 national and 3 international meetings of professional societies. Some of the major findings to date from Thunder Project II are outlined in the Table
according to their relationship to the multidimensionality of pain. The clinical implications of these findings and recommendations are also included.
|
Many other nurses and physicians deserve recognition for their work to improve treatment of pain in critically ill patients. A limitation of this article is the inability to recognize and thank each of them individually. For myself, I can identify many research roads to travel. For instance, I think that we should continue to investigate nonverbal indicators of pain, particularly pain behaviors, that can be used to validate pain in critically ill patients with communication barriers. Moreover, my selection as a Soros Faculty Scholar with Project on Death in America has provided support to my research team to investigate the particular challenges of assessing pain and other symptoms as well as providing palliative care for ICU patients at high risk of dying. And, as a side road, emergency department colleagues and I continue a series of studies on pain assessment and management in emergency departments, where similar challenges exist in providing patients’ comfort.
As a professor at UCSF, I am privileged to provide research assistance to students in doctoral and master’s degree programs and to consult with advanced practice nurses and physicians throughout the world, all of whom are advancing the science of pain assessment and management in critical care. I am convinced that through the work of other investigators and, I hope, my own continued efforts, research will continue to provide important insights on assessment and management of pain. In short, we do not have to rely on wizardry to improve the comfort of vulnerable, critically ill patients. Wizardry is being skilled in magic; researchers must rely on scientific advances to help lead us along the path.
Recently AACN President Connie Barden wrote, "Nurses are the highest authority when it comes to addressing pain and creating safe and humane environments. We must find solutions. We must listen and act from the inside out."43 I am in full agreement with her statement, and I would like to add how important it is that we nurses remain on our chosen career paths.
A Path Well Taken
On her way to the Emerald City to see the Wizard, Dorothy is helped by friends with special qualities: brains, heart, and courage. Walking my own path with others has been enlightening, enriching, and life essential. I am grateful to the many people who have guided me along my path. I feel privileged to be supported by my nursing and medical colleagues at UCSF, particularly Christine Miaskowski, my departmental chairperson. I am grateful for the research funding received from my university, professional organizations such as AACN, the Soros Foundation, and the National Institute for Nursing Research. Throughout my many years in nursing, my patients and students have been my greatest teachers, and the nurses with whom I work in the ICU providing care to patients have earned my utmost respect. These people have all been instrumental in helping me contribute to the goal that I think should be important to all of us: better pain management for critically ill patients.
To purchase reprints, contact The InnoVision Group, 101 Columbia, Aliso Viejo, CA 92656. Phone, (800) 809-2273 or (949) 362-2050 (ext 532); fax, (949) 362-2049; e-mail, reprints{at}aacn.org.
This article has been cited by other articles:
|
|
 |
|
  S. M. Carroll Silent, Slow Lifeworld: The Communication Experience of Nonvocal Ventilated Patients Qual Health Res, November 1, 2007; 17(9): 1165 - 1177. [Abstract] [PDF]
|
|
|
|
 |
|
 Y. Aissaoui, A. A. Zeggwagh, A. Zekraoui, K. Abidi, and R. Abouqal Validation of a Behavioral Pain Scale in Critically Ill, Sedated, and Mechanically Ventilated Patients Anesth. Analg., November 1, 2005; 101(5): 1470 - 1476. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
