Procedural Pain in the Adult Neurological Intensive Care Unit: A Retrospective Study Examining Arterial Line Insertion

Pain Management Nursing xxx (xxxx) xxx

Contents lists available at ScienceDirect

Pain Management Nursing journal homepage: www.painmanagementnursing.org

Original Article

Procedural Pain in the Adult Neurological Intensive Care Unit: A Retrospective Study Examining Arterial Line Insertion Kaylee Bray, DNP, MN, MSN *, Chris Winkelman, PhD y, Esther I. Bernhofer, PhD y, Jane F. Marek, DNP, MSN y * y

Louis Stokes Cleveland Veterans Affairs Medical Center, Cleveland, Ohio Frances Payne Bolton School of Nursing, Cleveland, Ohio

a r t i c l e i n f o

a b s t r a c t

Article history: Received 12 October 2018 Received in revised form 12 September 2019 Accepted 14 September 2019

Background: This was a retrospective chart review of procedural pain assessments and interventions during arterial catheter insertion in an adult neurological intensive care unit where patients with impaired consciousness are common. Overall, pain assessment was well documented (100%) by Registered Nurses, but not specific to arterial line insertion. Nurse practitioners commonly placed arterial lines and used local analgesia in over 75% of the documented procedures. Aims: The purpose of this study was to examine healthcare providers’ pain-related practices documented during arterial catheter insertion, one of the most painful procedures in a neurological intensive care unit. Secondary purposes were determining whether patient characteristics, procedure-related factors, or provider licensure were associated with pain assessment or procedural pain interventions. Design: A retrospective records review design was used. Methods: 120 electronic patient medical records were reviewed during a one-year period. Results: 100 charts met inclusion criteria. Nurses assessed all pain within 4 hours following the procedure in all charts but procedure-specific pain assessments were documented in 4% of charts. Painrelated interventions for arterial line insertion were local analgesic (76% of charts) and other procedure-specific interventions (10%). Significant associations occurred between procedure specific pain assessments and decreased number of insertion attempts (p ¼ .006) and between pain interventions and number of insertion attempts (p ¼ .003). No provider documented procedural pain assessment regarding arterial line insertion. Associations between patient characteristics and pain interventions were significant for patient ethnicity (F ¼ 8.967, p ¼ .007). Conclusions: Overall pain assessment was documented (100%) but not specific to arterial line insertion. Although arterial line insertion can be extremely painful, patients were rarely assessed for such pain by any clinician; 14% did not receive any preprocedural analgesia. Clinical Implications: The lack of procedural pain assessment in this vulnerable population indicates a need for increased pain management education for clinicians and further investigations to determine whether sufficient analgesia is provided to reduce procedural pain during arterial line insertion. Published by Elsevier Inc. on behalf of the American Society for Pain Management Nursing

Critically ill patients commonly report pain, and that pain can be associated with several procedures performed in the intensive care unit ([ICU]; Barr et al., 2013a; Marsac & Funk, 2008). Often, critically ill patients are unable to participate in self-reporting pain due to endotracheal intubation, sedation, or an altered level of

Address correspondence to Kaylee Bray, DNP, MN, MSN, Louis Stokes Cleveland VAMC, 111(w), 10701 East Boulevard, Cleveland, OH 44106. E-mail address: [email protected] (K. Bray).

consciousness (Barr et al., 2013a). It is the healthcare professional's responsibility to anticipate procedural pain and to provide interventions to reduce pain during procedures (Puntillo et al., 2012). Procedural pain has been extensively studied in a variety of patient populations, including children, patients with chronic pain, patients at end of life, and patients postsurgery (Czarnecki et al., 2011; Puntillo et al., 2001; Sigakis & Bittner, 2015). However, there is a lack of studies focusing on the adult intensive care population, and more specifically, the adult neurological intensive care population. The ability to self-report and express oneself verbally is

https://doi.org/10.1016/j.pmn.2019.09.003 1524-9042/Published by Elsevier Inc. on behalf of the American Society for Pain Management Nursing

Please cite this article as: Bray, K et al., Procedural Pain in the Adult Neurological Intensive Care Unit: A Retrospective Study Examining Arterial Line Insertion, Pain Management Nursing, https://doi.org/10.1016/j.pmn.2019.09.003

2

K. Bray et al. / Pain Management Nursing xxx (xxxx) xxx

often compromised in patients with central neurological impairment. The critically ill neurological intensive care adult population is particularly vulnerable to underassessment and undertreatment for procedural pain given the variety and frequency of procedures completed in this environment. Nurses and advanced-practice providers are in a unique position to assess and manage procedural pain in an ICU setting. A common procedure in the neurological intensive care unit is insertion of arterial catheters. Although not every patient will experience an arterial catheter placement, anecdotal and patient record data at the institution where this investigation was conducted indicates that as many as 60% of newly admitted patients will undergo arterial catheter insertion or insertion attempts. Arterial catheter insertion has been identified as one of the most painful procedures in the ICU (Barr et al., 2013b; Morrison et al., 1998; Nelson et al., 2001; Puntillo et al., 2014; Turner, Briggs, Springhorn, & Potgieter, 1990). Puntillo, Max, Chaize, Chanques, and Azoulay (2016) found that pain experienced while in the ICU had substantial and continuous physical and psychological effects. Procedural pain from arterial line insertion warrants attention and adequate pain assessment, pain management, and the associated documentation. Little current information exists about procedurerelated pain from arterial catheter insertion in adults who are hospitalized in the adult ICU, and there are no reports specific to critically ill neurological intensive care patients. The purpose of this study was to examine pain-related practices documented by clinicians in a neurological intensive care unit for the common procedure of arterial catheter insertion. A secondary purpose was to examine whether patient characteristics, procedure-related factors, or healthcare provider licensure were associated with either pain assessment or use of pain interventions during arterial catheter insertion in a neurological intensive care unit. Literature Review Pain experienced in the ICU is not an isolated event and can have long-term effects on patients’ perceived ICU pain and distress. Optimal pain management is considered by the Joint Commission to be a measure of competent care in hospital settings (Baker, 2016). The pain experienced from procedures performed in the ICU is underresearched and underestimated despite ICU pain having been identified as a psychological and physiological stressor (Gelinas & Johnston, 2007; Granja et al., 2005; Puntillo et al., 2001, 2004). Many published studies are older than 5 years, and their results imply procedural pain is not well managed, suggesting great reason to perform the present study. Puntillo and Colleagues (2001, 2004) completed the Thunder Project II, the largest published study to date on procedural pain in ICU patients. This comprehensive, descriptive study consisted of 6,210 verbal patients who rated pain intensity and procedural distress during six procedures identified as routinely performed in the participating ICUs. A major limitation of the Thunder Project II was that patients who were sedated or unable to communicate were excluded. Exclusion of nonverbal patients in studies of pain assessment and management is a significant gap in data. It is common that a patient in the neurological intensive care unit is not able to communicate verbally or physically owing to intubation or loss of brainstem reflexes (Wijdicks, Bamlet, Maramattom, Manno, & McClelland, 2005). A second study supported the finding that procedural pain in the ICU is both common and distressing (Puntillo et al., 2016). In the results, the authors reported that patients recalled procedural pain intensity and distress scores to be greater after discharge compared with what they reported while in the ICU. A significant and positive

correlation was found between degree of procedural pain intensity and days between discharge and interview. Puntillo acknowledged that “watershed” timedinitially described by Geng, Chen, Lam, and Zheng (2013) as a reconstructed memory based on subsequent knowledge acquisition that can develop 3 to 7 weeks after an experiencedcould have been a factor in the finding. The sample was comprised of verbal residents of France. Findings that perceptions of procedural pain increased up to 9 months after discharge, and that the pain experienced while in the ICU had substantial and continuous physical and psychological effects, makes continued research of procedural pain in the ICU imperative. Managing pain is thought to reduce post-ICU syndromes affecting mood, cognition, and functional impairment (Garrett, 2016). Puntillo and colleagues (2014) also investigated determinates of procedural pain intensity in the ICU. The three most painful procedures were (1) chest tube removal, (2) wound drain removal, and (3) arterial line insertions. The authors reported a doubling of selfreported pain scores during procedures compared to preprocedural pain scores. However, patients experiencing sedation and delirium were excluded from the sample. With the development and establishment of nonverbal pain assessment tools useful in sedated, delirious, and unconscious patients, it is time to revisit the patient experience of procedural pain in ICU. Nurse Practitioners and Pain Management in the ICU This study sought to investigate whether Advanced Practice Registered Nurses (APRNs) make unique contributions to mitigation of pain with arterial line insertion. None of the studies reviewed about procedural pain specifically examined APRN practice. APRNs providing direct care in the ICU is a relatively new phenomenon, and understanding their approach to mitigating procedural pain is crucial (Landsperger, Semler, Wang, Byrne, & Wheeler, 2016; McRee & Reed, 2016; Simone, McComiskey, & Andersen, 2016). Because APRNs are also registered nurses (RNs), they have the skill sets of both an RN and an APRN. Nurses play a vital role in assessing and determining a patient's pain level and adequately treating the pain (McCaffery, Pogue-Geile, Ferrell, Petro, & Manuck, 2002). Nurses often demonstrate inadequate knowledge of pain assessment and interventions, although this finding is not consistent (Al-Shaer, Hill, & Anderson, 2011; Thompson, Andrews, & Christ-Libertin, 2012). Although the current study did not examine either RN or APRN knowledge of pain assessment and interventions, it did examine their documentation related to procedural pain assessment and interventions. As prescribing providers, APRNs play a vital role in patient pain management. A study by Goldie, Prodan-Bhalla, and Mackay (2012) examined postoperative cardiac surgery care by APRNs compared with medical doctors (MDs). The researchers found that the nurse practitionereled patient teams had significantly higher scores for teaching, answering questions, listening skills, and pain management. Another report supports the link between APRN involvement and positive pain management outcomes by detailing promptness of assessment and therapeutic intervention, in addition to an absence of clinical incidents (Schoenwald, 2011). Research Questions This study sought to answer two research questions. The first was “Among adult patients in the Neurological Intensive Care Unit, was procedural pain assessed/evaluated before or after arterial catheter insertion?” Three sub-questions included who documented the pain assessment, what tool was used, and what was the average pain intensity score. The second research question was

Please cite this article as: Bray, K et al., Procedural Pain in the Adult Neurological Intensive Care Unit: A Retrospective Study Examining Arterial Line Insertion, Pain Management Nursing, https://doi.org/10.1016/j.pmn.2019.09.003

K. Bray et al. / Pain Management Nursing xxx (xxxx) xxx

“What pain prevention interventions or pain treatments were performed during arterial catheter insertion?”

Secondary Questions Three secondary research questions investigated whether procedural characteristics, patient characteristics, or provider licensure were associated with pain assessments or interventions. The first question was “Are procedural characteristics (e.g., site of placement, urgency of the procedure, occurrence of complications related to catheter placement) associated with assessment or interventions (or lack of pain assessment and lack of pain management)?” The second question was “Are patient characteristics (e.g., age, ethnicity, sex, admitting diagnosis, day of ICU stay) associated with assessment or interventions (or lack of pain assessment and lack of pain management)?” The third question was “Is provider licensure (APRN, physician assistant [PA], medical doctor [MD], Doctor of Osteopathy [DO], MD fellow, or MD resident) associated with assessment and interventions (or lack of pain assessment and lack of pain management)?”

Theoretical Framework The International Association for the Study of Pain (Bogduk & Merskey, 2002, p. 210) defines pain as an “unpleasant sensory and emotional experience associated with actual or potential tissue damage, or described in terms of such damage.” Acute pain is a perception that is initiated by the stimulation of peripheral nociceptors and transmitted to the sensory cortex by specialized pain fiber systems (Puntillo et al., 2001). Puntillo et al. (2001) further defines procedural pain as “the unpleasant sensory and emotional experience that arises from actual or potential tissue damage associated with diagnostic or treatment procedures” (p. 240). Patients report significant pain and distress from pain while in the ICU (Siffleet, Young, Nikoletti, & Shaw, 2007). The Synergy Model The American Society for Pain Management Nursing recommends that healthcare professionals advocate and intervene to provide optimal comfort management before, during, and after procedures, based on the needs of the patient, the setting, and the situation (Czarnecki et al., 2011). Combining the actions of the healthcare provider and the needs of the patient is the concept behind the American Association of Critical-Care Nurses (AACN) Synergy Model for Patient Care (Hardin & Kaplow, 2017). The core concept of the Synergy Model is that the needs or characteristics of patients and families should be matched to the characteristics or competencies of the healthcare provider. Optimum patient outcomes occur when these two components synergize. Synergy occurs when healthcare provider competencies are driven by the needs of the patient, the clinical unit, or the system. The Synergy Model for Patient Care provides an important rationale for studying who assesses and manages procedural pain.

Method Design This study was a retrospective chart review that examined documentation of pain and pain interventions during arterial catheter insertion in the Neurological ICU.

3

Ethics This study was approved by the setting's institutional review board. Linkage data were kept in a separate electronic file and destroyed once preliminary analyses were completed. All data were saved on a Kingston IronKey D300 USB, a hardware-encrypted USB drive, to ensure data protection. Setting Data were collected from charts of patients admitted to a quaternary-care academic urban medical center. The Neurological ICU was a 24-bed unit with five attending MDs, three MD fellows, and two to four rotating resident MDs each month. There were seven APRNs on staff who managed seven patients during each shift, including initiating procedures such as arterial line insertion and interventions such as pain management. There were no physician assistants (PAs) providing direct care during the study period. Sample The study sample included charts of patients who underwent an arterial catheter insertion between October 1, 2016, and October 1, 2017. Only adults over 21 years of age were included. There were no exclusion criteria based on age, sex, or race. However, patients with charts documenting a second or third arterial line insertion on a single patient were excluded, as multiple attempts may require different pain-related interventions. Charts with missing data for 20 or more variables of interest were excluded. Charts were consecutively selected from a list of 348 charts categorized by last name and service date provided by the billing department. Sample Size This was a descriptive study, so sample size would not typically be calculated. However, a sample size of 100 was determined to be adequate for identifying issues and concerns, based on a preliminary review of 10 charts per variable of interest (Peduzzi, Concato, Kemper, Holford, & Feinstein, 1996). Oversampling to 120 charts was planned to allow for an incorrectly coded procedure or charts with more than 20% missing data points. Variables and Instrument Although most variables (e.g., age, sex, ethnicity, admitting diagnosis, day of ICU when arterial line was placed, number of arterial catheter insertion attempts) were self-explanatory, several variables were defined with greater detail prior to data collection. Sites for placement of arterial lines included radial, axillary, and femoral arteries. A Glasgow Coma Score (GCS) of <8 indicates coma, and a Richmond Agitation-Sedation Scale (RASS) score of -5 to -1 indicates reduced ability to interact with the environment (Sessler et al., 2002). Urgency of the procedure was coded as emergent if consent was not obtained before the procedure, based on the assumption that immediate action was needed. The dependent variable of pain intervention was categorized into nonpharmacological (e.g., distraction or positioning) and pharmacological. Pharmacological pain interventions were further categorized as local/nonopioid analgesic, systemic nonopioid analgesic, systemic opioid analgesic, sedative, or a combination. The rationale for including sedatives and benzodiazepine as pain interventions was that use of these medications can interfere with forming a memory of pain, leading to fewer outcomes of reported distress a

Please cite this article as: Bray, K et al., Procedural Pain in the Adult Neurological Intensive Care Unit: A Retrospective Study Examining Arterial Line Insertion, Pain Management Nursing, https://doi.org/10.1016/j.pmn.2019.09.003

4

K. Bray et al. / Pain Management Nursing xxx (xxxx) xxx

month after painful procedures; continuous or intermittent sedation is thus a factor that could alter procedural pain perception. All data were entered into a spreadsheet developed by the principal investigator. Coauthors reviewed the spreadsheet for accuracy and completeness prior to data collection. Procedure The institution's billing department identified all medical records with the Current Procedural Terminology code for arterial catheter insertion performed in the Neuro ICU during study dates. Data were collected from electronic patient medical records by a single trained data collector over 6 weeks. Charts were reviewed to identify recorded assessments from the time of the procedure and any routine pain assessment conducted within 4 hours after the procedure. Documentation of interventions, if performed, at the time of insertion and within 4 hours after the procedure of arterial catheter insertion were collected to detail pain interventions. The sources of data included the ICU flowsheet, the medication administration records (MAR), the procedural note, and narrative notes at the time of arterial line insertion. The electronic ICU flowsheet typically documents pain using a dropdown menu with options to record pain tool, score, and interventions. The MAR section of the patient's chart includes administered drugs and comments related to pain assessment or reason for administration. The procedure note uses dropdown choices for each component of a typical procedural note. Narrative notes and MAR comments were examined for interventions such as the use of distraction, positioning, local anesthetics, and oral or IV analgesics and sedatives.

Table 1 Summary of Pain Interventions Intervention

Documentation of Pharmacological Pain Management Yes No Pharmacological Pain Management Used Local Analgesic Lidocaine 1% Lidocaine 2% Opioid Analgesic Sedative Intermittent Midazolam Continuous Propofol Continuous Midazolam Local Analgesic and Opioid Analgesic Local Analgesic, Opioid Analgesic, and Sedative Local Analgesic and Sedative Documented Pain Interventions in RN Flowsheet Medication Reposition Medication and Repositioning Medication, Repositioning, and Relaxation Repositioning and Relaxation Repositioning, Relaxation, and Education Not Documented

Results and Discussion Sample A total of 120 charts were examined. There were 20 occurrences of a second or third arterial catheter placement among charts randomized for inclusion, and these 20 records were not included in the final analysis. The sample was a total of 100 unique patient charts with an initial arterial line placement. The patient characteristics documented in the charts are detailed in Table 2. The sample consisted of an equal distribution of sexess (50% male, 50% female); ethnicity distribution was majority white (73% White, 25% African-American, and 2% Other [undocumented]). Fifty-one percent of patients were intubated, and 37% had a low level of consciousness, as indicated by a GCS <8. Decreased consciousness is further supported in the 55% of patients who had a RASS score indicating low interaction with the environment.

88 (88%) 12 (12%) 67 73 3 4 8 1 7 4 5 1 3

(67%) (73%) (3%) (4%) (8%) (1%) (7%) (4%) (5%) (1%) (3%)

6 3 3 9 2 1 2

(6%) (3%) (3%) (9%) (2%) (1%) (2%)

Table 2 Sample Summary Data Patient Characteristic

Mean (Standard Deviation)

Age Sex Male Female Ethnicity White African-American Other Day of ICU Stay (when arterial atheter was placed)

Mean 58.5 (14.12)

Analyses All data were analyzed with SPSS (version 25, https://www.ibm. com). Descriptive data were summarized with means and medians (if continuous) and frequencies (if nominal or categorical) as appropriate. Planned analysis for research question 1 was descriptive, including frequency and means. For research question 2, pain interventions were detailed with summary statistics. Secondary research questions were examined with logistic regression (with the binary dependent variable “assessment did/did not occur”) and ANOVA (with the dependent variable of multiple pain interventions, as described in Table 1). Associations between provider and assessment or intervention were planned for testing with chi-square associations.

Frequencies Number (Percent)

Admitting Diagnosis Subarachnoid Hemorrhage Intracranial Hemorrhage Ischemic Stroke Seizure Other (Cavernous malformation, Symptomatic aneurysm, Vasospasm, etc.) Glasgow Coma Score (GCS) Non-intubated participants Intubated participants >8/8T* <8/8T* Richmond Agitation and Sedation Score (RASS) (99; 1 participant did not have RASS documented)

Frequency Number (Percent)

50 (50%) 50 (50%)

Mean 2.36 (3.5)

73 (73%) 25 (25%) 2 (2%) Day 1 ¼ 52 (52%) Day 2 ¼ 28 (28%) Day 3 ¼ 9 (9%) Day 4 ¼ 3 (3%) Days 12, 13, 15, or 29 ¼ 1 each (4%) 25 34 17 10 14

14 (1.4) Range 10-15 7T (2.7) Range 3T-13T 1 (4.0)

(25%) (34%) (17%) (10%) (14%)

49 (49%) 51 (51%) 63 (63%) 37 (37%) þ4, þ5 ¼ 0 þ3 ¼ 1 þ2 ¼ 3 þ1 ¼ 7 0 ¼ 33 1 ¼ 14 2 ¼ 9 3 ¼ 4 4 ¼ 16 5 ¼ 12

* 7T, 8T in the Glasgow coma Score indicate the patient is intubated and not able to complete the verbal scale.

Please cite this article as: Bray, K et al., Procedural Pain in the Adult Neurological Intensive Care Unit: A Retrospective Study Examining Arterial Line Insertion, Pain Management Nursing, https://doi.org/10.1016/j.pmn.2019.09.003

K. Bray et al. / Pain Management Nursing xxx (xxxx) xxx Table 3 Summary of Procedural Factors Procedure Characteristic Urgency of Arterial Line Placement Consent obtained; non-emergent No consent; emergency Procedure Site Radial Axillary Licensure APRN Staff Physician Fellow Resident Complications During Procedure None recorded

Frequencies Number (Percent) 78 (78%) 22 (22%) 97 (97%) 3 (3%) 44 10 25 21

(44%) (10%) (25%) (21%)

100 (100%)

Arterial Catheter Placement The findings for the procedural variables are detailed in Table 3. APRNs placed 44% (n ¼ 44) of arterial catheters, whereas 56% (n ¼ 56) were placed by physicians (MD staff, fellows, and residents). No PAs or DOs placed arterial lines in this sample, so those licensures were not used in analyses. No femoral lines were inserted in these 100 participants, so this site placement was excluded from analyses. No complications were documented in the reviewed charts, so this variable was not included in analyses. Q1a. Is procedural pain assessed during or after arterial catheter insertion? A total of four charts (4%) had a procedure-specific assessment of pain related to arterial line insertion. All four assessments were documented by an RN. In addition to the procedural pain assessments, routine pain assessments were conducted within 4 hours of the arterial line placement in 96% of the charts. All routine pain assessments were documented by an RN. Over half of these routine pain assessments were assessed for pain with a nonverbal measure (58%), and this assessment was not specific to procedural pain.

Q1a Discussion Four percent of charts had assessments documented that were specific to pain management for arterial line insertion. This finding is similar to Puntillo et al's (2009) earlier reports, in which procedural pain was rarely assessed. No existing institutional or professional guidelines mandate documentation of procedural pain. If indeed memories of procedural pain persist and can be recalled as more severe over time, as Puntillo et al. (2016) suggest, then perhaps we should focus on education of providers and documentation by the proceduralist to identify and mitigate procedural pain. Q1b. What tool was used? Of the four charts that documented a pain assessment specifically for the procedure, one used the numeric pain score (i.e., 0-10 oral self-report) and three used the Nonverbal Pain Assessment Tool (NPAT; Klein et al., 2010). The NPAT, developed by Klein, Dumpe, Katz, and Bena (2010), is a behavioral pain assessment tool that was developed for the nonverbal adult patient in the ICU and was in common use during the study period in this setting. It has a range of 0-10, similar to the verbal scale for pain.

5

For the 96 charts in which routine pain assessments occurred within 4 hours after the arterial line insertion, a numeric pain score was used in 41 assessments, and the NPAT was used in 55 charts. For all charts assessed (N ¼ 100), a numeric pain score was used in 42 assessments and the NPAT was used in 58 assessments.

Q1b Discussion When procedural pain was documented, both verbal and nonverbal tools were used. The regular use of nonverbal pain assessments reflects a change in practice since Puntillo and colleagues’ reports of procedural pain (Puntillo et al., 2004, 2014, 2016). Objective, nonverbal measures for nonverbal patients are valued and used in clinical settings (Azevado-Santos & DeSantana, 2018). The current study has new information about the routine assessment of pain in adults with critical illness. Pain was documented in 100% of the charts reviewed (96% routine assessments and 4% specific to procedural pain), and 58% of documented pain scores used the NPAT. It was encouraging to see pain regularly assessed, even among patients with an altered level of consciousness or endotracheal intubation. The clinicians placing the catheter consistently used a dropdown menu to document the procedure, including the patient response to the procedure. The provider performing the procedure selected “patient tolerated procedure well” in 98% of the charts and “patient tolerated procedure fairly well” in 2% of the charts. The procedural note dropdown menu did not include an assessment of pain, but did include a dropdown menu for “Anesthesia/Sedation.” Because pain is subjective, self-report assessments are the common approach to pain assessment. The most common selfreport assessment used in the ICU setting is the Numerical Rating Scale (Sigakis & Bittner, 2015). Behavioral assessments are also recognized as valid and reliable measures of patient pain in ICU settings when patients are unable to verbalize, or have cognitive impairment that interferes with their ability to use a numerical scale to report pain. One valid and reliable tool for nonverbal patients to report pain is the Nonverbal Pain Assessment Tool (Klein et al., 2010). The inability to communicate pain experiences does not negate the possibility that the person is experiencing pain (Puntillo et al., 2009). The regular use of a nonverbal pain assessment tool was a new finding not previously reported in procedural pain research by Puntillo and colleagues (2001, 2009, 2014). Q1c. What was the average pain intensity score? There were too few data to report an average of procedural pain. Three of the four charts with documentation of procedural pain recorded 0 of 10 pain score. The fourth documentation of pain during a procedure detailed 10 of 10 back pain occurring at the time of the procedure. Further investigation indicated that back pain was a recurrent complaint for this individual and may not have been related to the procedure, but simply documented at the same time as the procedure occurred. The mean value of the routine pain score recorded in the 4 hours after arterial line insertion was 1.61, and the overall mean for all charts (n ¼ 100) was 1.65. However, 19% of all participants reported a pain score equal to or greater than 5.

Q1c Discussion Previous studies show that patients report significant pain and related distress while in the ICU (Siffleet et al., 2007). Results in this

Please cite this article as: Bray, K et al., Procedural Pain in the Adult Neurological Intensive Care Unit: A Retrospective Study Examining Arterial Line Insertion, Pain Management Nursing, https://doi.org/10.1016/j.pmn.2019.09.003

6

K. Bray et al. / Pain Management Nursing xxx (xxxx) xxx

study included a finding of a low average pain score of only 1.65, combining numeric pain scores and NPAT scores. Although the pain intensity score had a low average value, 19 documented pain scores (19%) were greater than 5 of 10. Q2. What pain prevention interventions or pain treatments are performed during arterial catheter insertion? The most common intervention was use of local analgesia by the clinician placing the catheter. Isolated opioid administration occurred in 4% and continuation of intravenous sedatives was documented in 12% of charts reviewed. A local analgesic was used for 76 patients. The clinician placing the arterial catheter used local analgesia alone (n ¼ 67) or in combination with an opioid (n ¼ 5), a sedative (n ¼ 3), or an opioid and a sedative (n ¼ 1). When an opioid was used, it was fentanyl and administered as an intravenous bolus. The most common local analgesic was 1% lidocaine (used in 96%, n ¼ 73), and 2% lidocaine was used for the remainder (4%, n ¼ 3). Additional pain interventions documented in the RN Flowsheet included relaxation, repositioning, and education. There were 10 occurrences of pharmacological intervention documented at the time of a procedure in response to pain related to the procedure. Documentation occurred in the procedure notes, RN notes, or MAR comment. In these 10 charts, the numeric pain score (n ¼ 2) and the NPAT (n ¼ 8) were used to document response to procedural pain interventions. A routine pain intensity score greater than 1 of 10 was documented in 27 charts (27%) around the time of the arterial line insertion, and interventions were performed in 24 of the 27 charts. However, the pain assessment and interventions were deemed not specific to the procedure based on a review of narrative and MAR notes, and these data are not included in the discussion.

Q2 Discussion The Joint Commission considers optimal pain management a measure of competent care in the hospital setting (Baker, 2016). Previous reports indicate that fewer than 20% of patients receive opiates before procedures (Czarnecki et al., 2011; Puntillo et al., 2012); this is similar to our finding that 21% of the sample received an opioid or sedative to manage procedural pain. Although not all procedures necessitate opioid administration for comfort, further investigation into patient preferences around the painful procedure of arterial line insertion is warranted. Local analgesic use to prevent pain associated with arterial line placement was documented in 76% of charts. This is a new finding. Although use of local subcutaneous lidocaine is the common practice for pain management during arterial catheter insertion, there are no reports examining the efficiency or patient-reported effectiveness of this practice. Further studies are needed to explore the use of topical agents (e.g., topical lidocaine cream, lidocaine patch) or nonpharmacological interventions (e.g., vibration, cold, distraction, or music) to manage arterial line placement pain. It is also not clear why 24% did not receive local analgesic. There might be an association between “emergent” status and lack of lidocaine administration, but this finding did not emerge in our results. Perhaps the surrogate of using lack of consent for “emergent” status may not be valid. A prospective study may more clearly categorize emergent procedures. Nurses are in a unique position to assess and manage procedural pain in an ICU setting, and this study found that RNs documented all (100%) of the pain assessments in this study. If other healthcare providers should be documenting pain, particularly around

procedures, it may be that the procedural note and dropdown menus around procedures should be adjusted to include a specific report of pain assessment and the response reported. It may be that provider documentation that “patient tolerated procedure well” or “patient tolerated procedure fairly well” indicates that pain was not distressing. A prospective study would be better able to discern patient perception of pain during arterial catheter placement. Unlike the routine (i.e., every 4 hours) assessment for the presence of pain in an inpatient setting mandated by the Joint Commission, assessment of procedural pain is neither mandated nor a component of the procedure note. Arterial catheter placement was one of the top three most painful reported procedures in previous reports. No existing reports address how patients prefer this procedure pain to be managed (Siffleet et al., 2007). In this study, 76% of patients received lidocaine for local analgesia before arterial line insertion. It may be that the use of subcutaneous lidocaine is sufficient to manage radial or axillary arterial line insertion; further investigation is warranted in a prospective study with nonverbal patients. A prospective study could examine patient baseline (immediately before), procedural (during or immediately following), and postprocedure pain (in the 10-30 minutes after procedure completion) to better understand whether local analgesia is sufficient. Q3. Are procedural characteristics (site for placement, urgency of the procedure, occurrence of complications related to catheter placement) associated with assessment or interventions (or lack of pain assessment and lack of pain management)? Logistic regression analyses were conducted to examine the association between procedural pain assessed and the procedural variables. A test of the full model was not statistically significant, indicating that the procedure-related variables did not distinguish between the occurrence of procedural pain assessment and the lack of such assessment (chi-square ¼ 2.409, p ¼ .492). The Nagelkerke's R2 of .083 indicates no relationship between the variables; multicollinearity was not an issue. The 95% confidence intervals support a finding of no significance, as all intervals included 1 in the analyses. The logistic regression statistical model predicted less than 9% of variation in assessments for procedural pain (Exp[B] ¼ 0.865). It may be that an insufficient number of charts were reviewed for the logistic regression model, and that the finding of nonsignificance is spurious (Long, 1997). ANOVA was used to determine associations between procedural pain interventions and the three procedural variables. Documented procedural pain assessment was significantly associated with fewer numbers of attempts (F ¼ 7.422, p ¼ .006). Neither site for placement (i.e., radial or axillary; F ¼ 0.344, p ¼ .558) nor urgency of the procedure (emergent versus nonemergent; F ¼ 0.026, p ¼ .872) was significantly associated with procedural pain assessment in the logistic regression analysis.

Q3 Discussion Although it was uncommon for procedural pain assessment to be documented, routine pain assessment was completed in the 4 hours after the procedure by an RN in 96% of the charts reviewed. This is similar to findings by Thompson et al. (2012), who reported that RNs are competent in practicing safe and effective pain management during patient care. The overall low average pain intensity score may be one reason that pain was not specifically addressed during procedures. The low pain intensity score may be the result of reduced verbal reporting and facial and body relaxation associated with the use of continuous IV sedation

Please cite this article as: Bray, K et al., Procedural Pain in the Adult Neurological Intensive Care Unit: A Retrospective Study Examining Arterial Line Insertion, Pain Management Nursing, https://doi.org/10.1016/j.pmn.2019.09.003

K. Bray et al. / Pain Management Nursing xxx (xxxx) xxx

(11% of patients; Table 1). It may be that when continuous (deep) sedation for seizure management or ventilator noncompliance is used, patients are less likely to demonstrate pain, even nonverbally. The overall low pain score could also be a result of the NPAT not being an adequate pain assessment tool for the neurological ICU population. The three charts that indicated no intervention despite a positive finding of pain may reflect the challenges of caring for patients with intractable pain, such as headache when blood is in the cerebrospinal fluid from either subarachnoid hemorrhage or vasospasm. It may be that this snapshot caught a period of time when pain was not well-managed due to the need for neurological monitoring to provide a surgical intervention decision, limiting the use of opioids. Patients with prolonged pain may have unique responses to procedural pain (Puntillo et al., 2014). This study did not examine pain over time. Q4. Are patient characteristics (age, ethnicity, sex, admitting diagnosis, day of ICU stay) associated with assessment or interventions (or lack of pain assessment and lack of pain management)? The result of logistic regression examining the association of procedural pain assessment and patient characteristics was not significant (chi square ¼ 3.29, p ¼ .663). The Nagelkerke's R2 value was .112, indicating a weak relationship (<12%) between the dependent variables, but logistic regression is robust to this small value for associations. Correlations between patient characteristics were small (<.15), and confidence intervals did not support significance. The Exp(B) value of 0.9 indicated that this model did not predict association between procedural pain assessment and the patient characteristics included in analysis. ANOVA results examining associations between pain interventions and patient characteristics were significant for the patient characteristic of ethnicity (F ¼ 8.967, p ¼ .007). There were no associations between documented procedural pain interventions and other patient characteristics of age, sex, admitting diagnosis, or day of ICU stay (F < 8.9, p < .3).

7

Q5. Is provider licensure (i.e., APRN, MD, fellow, and resident) associated with assessment and interventions (or lack of pain assessment and lack of pain management)? There were no PAs or DOs placing catheters in the charts reviewed, so these two licensure categories were eliminated from analyses. No significant relationships were found between provider licensure and either assessment (F ¼ 2.446, p ¼ .132) or interventions (F ¼ 0.928, p ¼ .345). Grouping MDs together did not change the result of insignificance.

Q5 Discussion The core concept of the Synergy Model is that the needs or characteristics of patients and families should be matched to the characteristics or competencies of the healthcare provider. This study found no relationship between license of the person performing the procedure and either assessment or interventions. This finding suggests that APRNs provide similar care (lacking assessment but providing local analgesia) as physicians. The licensed provider who performed the procedure of arterial line placement documented “tolerated procedure well” in 98 charts and “tolerated procedure fairly well” in 2 charts, but 27 patients reported pain greater than 0 of 10 in routine assessments within 4 hours after the procedure. Ten subjects were given opioids during the procedure or within 4 hours following the procedure. Twelve of the 27 patients were on continuous sedation for indications other than procedural pain management. Further studies are needed to explore if the provider performing the procedure should consider pain assessment and management as essential to procedural care and documentation. For example, the procedure note in the electronic medical record could prompt a targeted assessment and detail the interventions used to ensure the proceduralist is involved in pain assessment and related interventions. Studies over time are also needed to better understand the long-term sequelae from procedural pain. This retrospective study examined only one procedure and only one period of time. Limitations of the Study

Q4 Discussion A healthcare provider has a moral obligation to act in the best interest of the patient while preventing undue harm; this includes treating patients with dignity, respecting the uniqueness of individuals, and supporting autonomy in choices such that individual views and experiences will be heard by their healthcare provider regardless of demographics (Czarnecki et al., 2011). Associations between pain interventions and patient characteristics were positively significant for ethnicity (F ¼ 8.967, p ¼ .007). These findings suggest that pain interventions coded as higher numbers (e.g., medication, relaxation, repositioning, education) were associated with African-American ethnicity. Bell et al. (2018) demonstrated significantly elevated pain sensitivity in AfricanAmerican patients compared to non-Hispanic white patients, as measured by the Pain Sensitivity Questionnaire (English version) and experimental and clinical pain intensity scores. The literature review identified racial disparities in the palliative care of patients with cancer and their impact on quality of life for AfricanAmerican women; the authors reported that although sexrelated influences on drug response have been documented, to date, no racial or ethnic differences in analgesic responsiveness have been reported (Payne, Medina, & Hampton, 2003). Future studies are needed to assess pain treatment disparities by identifying and standardizing differences in pain sensitivity.

A significant limitation of a retrospective study is that it relies on the accuracy of charted patient data and the assumption that all data of interest have been documented (Hess, 2004). A retrospective study design also does not control for bias, and there can be little randomization (Hess, 2004). However, this study sought to avoid bias by using a larger sample size of 100 charts and nonbiased, consecutive selection of those charts. Although a larger sample would build increased confidence that nonsignificance was not a spurious finding, this study did not seek to provide a causeand-effect analysis or provide a definitive study. It is a first step in understanding how procedural pain is assessed and managed in a population with impaired consciousness and, as such, it brings important new data to patient care in an ICU setting. This retrospective study examined only one procedure during a single period; future studies examining the long-term consequences of procedural pain in this population are needed. Conclusion The purpose of this study was to examine pain-related practices documented by clinicians in a neurological intensive care unit for the common, recurrent procedure of arterial catheter insertion. A retrospective chart review was conducted of 100 adult neurological ICU patients who underwent arterial catheter insertions. In this sample, many patients were not able to participate in self-report of

Please cite this article as: Bray, K et al., Procedural Pain in the Adult Neurological Intensive Care Unit: A Retrospective Study Examining Arterial Line Insertion, Pain Management Nursing, https://doi.org/10.1016/j.pmn.2019.09.003

8

K. Bray et al. / Pain Management Nursing xxx (xxxx) xxx

pain based on both averaged GCS and RASS values. Routine pain assessment occurred in 100% of the charts, as documented by RNs, many using the novel NPAT; this finding differs significantly from other reports in the literature about assessing pain in adult patients in an ICU. Patients were rarely assessed for arterial line insertion procedural pain by any clinician. Anticipation of procedural pain was managed with the use of a local analgesic for 76% of the patients. An association between pain interventions and ethnicity was significant and, instead of suggesting reduced pain care, suggested an increased number of pain interventions performed on people of African-American ethnicity. There were no significant relationships between provider licensure and either assessment or interventions. All pain assessments were documented by an RN. This chart review may guide future studies, including prospective investigations of pain assessment in nonverbal patients, to determine if local lidocaine is sufficient to reduce the experience of procedural pain during arterial line insertion. References Al-Shaer, D., Hill, P. D., & Anderson, M. A. (2011). Nurses’ knowledge and attitudes regarding pain assessment and intervention. Medsurg Nursing, 20(1), 7e11. Azevado-Santos, I. F., & DeSantana, J. M. (2018). Pain measurement techniques: spotlight on mechanically ventilated patients. Journal of Pain Research, 11, 2969e2980. Baker, D. W. (2016). Statement on pain management: understanding how Joint Commission standards address pain. Joint Commission Perspectives, 36(6), 10e12. Barr, J., Fraser, G. L., Puntillo, K., Ely, E. W., Gelinas, C., & Dasta, J. F. (2013a). Clinical practice guidelines for the management of pain, agitation, and delirium in adult patients in the intensive care unit. Critical Care Medicine, 41(1), 263e306. Barr, J., Fraser, G. L., Puntillo, K., Ely, E. W., Gelinas, C., & Dasta, J. F. (2013b). Clinical practice guidelines for the management of pain, agitation, and delirium in adult patients in the intensive care unit: executive summary. American Journal of Health-System Pharmacy, 70(1), 53e58. Bell, B. A., Ruscheweyh, R., Kelley, B. J., Ness, T. J., Vetter, T. R., & Sellers, A. B. (2018). Ethnic differences identified by pain sensitivity questionnaire correlate with clinical pain responses. Regional Anesthesia and Pain Medicine, 43(2), 200e204. Bogduk, N., & Merskey, H. (2002). Classification of chronic pain: Descriptions of chronic pain syndromes and definitions of pain terms (2nd ed.). Seattle, WA: International Association for the Study of Pain, Task Force on Taxonomy. Czarnecki, M. L., Turner, H. N., Collins, P. M., Doellman, D., Wrona, S., & Reynolds, J. (2011). Procedural pain management: a position statement with clinical practice recommendations. Pain Management Nursing, 12(2), 95e111. Garrett, K. M. (2016). Best practices for managing pain, sedation, and delirium in the mechanically ventilated patient. Critical Care Nursing Clinics of North America, 28(4), 437e450. Gelinas, C., & Johnston, C. (2007). Pain assessment in the critically ill ventilated adult: validation of the critical-care pain observation tool and physiologic indicators. Clinical Journal of Pain, 23(6), 497e505. Geng, X., Chen, Z., Lam, W., & Zheng, Q. (2013). Hedonic evaluation over short and long retention intervals: the mechanism of the peak-end Rule. Journal of Behavioral Decision Making, 26(3), 225e236. Goldie, C. L., Prodan-Bhalla, N., & Mackay, M. (2012). Nurse practitioners in postoperative cardiac surgery: are they effective? Canadian Journal of Cardiovascular Nursing, 22(4), 8e15. Granja, C., Lopes, A., Moreira, S., Dias, C., Costa-Pereira, A., Carneiro, A., & Group, J. S.. (2005). Patients’ recollections of experiences in the intensive care unit may affect their quality of life. Critical Care, 9(2), R96eR109. Hardin, S. R., & Kaplow, R. (2017). Synergy for clinical excellence: The AACN Synergy Model for Patient Care (2nd ed.). Burlington, MA: Jones & Bartlett Learning. Hess, D. R. (2004). Retrospective studies and chart reviews. Respiratory Care, 49(10), 1171e1174. Klein, D. G., Dumpe, M., Katz, E., & Bena, J. (2010). Pain assessment in the intensive care unit: development and psychometric testing of the nonverbal pain assessment tool. Heart and Lung, 39(6), 521e528.

Landsperger, J. S., Semler, M. W., Wang, L., Byrne, D. W., & Wheeler, A. P. (2016). Outcomes of nurse practitioner-delivered critical care: a prospective cohort study. Chest, 149(5), 1146e1154. Long, J. S. (1997). Regression models for categorical and limited dependent variables. Thousand Oaks, CA: Sage Publications. Marsac, M. L., & Funk, J. B. (2008). Relationships among psychological functioning, dental anxiety, pain perception, and coping in children and adolescents. Journal of Dentistry for Children, 75(3), 243e251. McCaffery, J. M., Pogue-Geile, M. F., Ferrell, R. E., Petro, N., & Manuck, S. B. (2002). Variability within alpha- and beta-adrenoreceptor genes as a predictor of cardiovascular function at rest and in response to mental challenge. Journal of Hypertension, 20(6), 1105e1114. McRee, L., & Reed, P. G. (2016). Nurse practitioners: knowledge, skills, and leadership for the end-of-life conversation in intensive care. Nursing Science Quarterly, 29(1), 78e81. Morrison, R. S., Ahronheim, J. C., Morrison, G. R., Darling, E., Baskin, S. A., Morris, J., & Meier, D. E. (1998). Pain and discomfort associated with common hospital procedures and experiences. Journal of Pain Symptom Management, 15(2), 91e101. Nelson, J. E., Meier, D. E., Oei, E. J., Nierman, D. M., Senzel, R. S., Manfredi, P. L., & Morrison, R. S. (2001). Self-reported symptom experience of critically ill cancer patients receiving intensive care. Critical Care Medicine, 29(2), 277e282. Payne, R., Medina, E., & Hampton, J. W. (2003). Quality of life concerns in patients with breast cancer: evidence for disparity of outcomes and experiences in pain management and palliative care among African-American women. Cancer, 97(1 Suppl), 311e317. Peduzzi, P., Concato, J., Kemper, E., Holford, T. R., & Feinstein, A. R. (1996). A simulation study of the number of events per variable in logistic regression analysis. Journal of Clinical Epidemiology, 49(12), 1373e1379. Puntillo, K., Pasero, C., Li, D., Mularski, R. A., Grap, M. J., Erstad, B. L., & Sessler, C. N. (2009). Evaluation of pain in ICU patients. Chest, 135(4), 1069e1074. Puntillo, K. A., Max, A., Chaize, M., Chanques, G., & Azoulay, E. (2016). Patient recollection of ICU procedural pain and post ICU burden: the memory study. Critical Care Medicine, 44(11), 1988e1995. Puntillo, K. A., Max, A., Timsit, J. F., Vignoud, L., Chanques, G., Robleda, G., & Azoulay, E. (2014). Determinants of procedural pain intensity in the intensive care unit. The Europain study. American Journal of Respiratory Critical Care Medicine, 189(1), 39e47. Puntillo, K. A., Morris, A. B., Thompson, C. L., Stanik-Hutt, J., White, C. A., & Wild, L. R. (2004). Pain behaviors observed during six common procedures: results from Thunder Project II. Critical Care Medicine, 32(2), 421e427. Puntillo, K. A., Neuhaus, J., Arai, S., Paul, S. M., Gropper, M. A., Cohen, N. H., & Miaskowski, C. (2012). Challenge of assessing symptoms in seriously ill intensive care unit patients: can proxy reporters help? Critical Care Medicine, 40(10), 2760e2767. Puntillo, K. A., White, C., Morris, A. B., Perdue, S. T., Stanik-Hutt, J., Thompson, C. L., & Wild, L. R. (2001). Patients’ perceptions and responses to procedural pain: results from Thunder Project II. American Journal of Critical Care, 10(4), 238e251. Schoenwald, A. V. (2011). Two hundred days of nurse practitioner prescribing and role development: a case study report from a hospital-based acute pain management team. Australian Health Review, 35(4), 444e447. Sessler, C. N., Gosnell, M. S., Grap, M. J., Brophy, G. M., O’Neal, P. V., Keane, K. A., Tesoro, E. P., & Elswick, R. K. (2002). The Richmond Agitation-Sedation Scale: validity and reliability in adult intensive care unit patients. American Journal of Respiratory Critical Care Medicine, 166(10), 1338e1344. Siffleet, J., Young, J., Nikoletti, S., & Shaw, T. (2007). Patients’ self-report of procedural pain in the intensive care unit. Journal of Clinical Nursing, 16(11), 2142e2148. Sigakis, M. J., & Bittner, E. A. (2015). Ten myths and misconceptions regarding pain management in the ICU. Critical Care Medicine, 43(11), 2468e2478. Simone, S., McComiskey, C. A., & Andersen, B. (2016). Integrating nurse practitioners into intensive care units. Critical Care Nursing, 36(6), 59e69. Thompson, E. M., Andrews, D. D., & Christ-Libertin, C. (2012). Efficacy and safety of procedural sedation and analgesia for burn wound care. Journal of Burn Care Research, 33(4), 504e509. Turner, J. S., Briggs, S. J., Springhorn, H. E., & Potgieter, P. D. (1990). Patients’ recollection of intensive care unit experience. Critical Care Medicine, 18(9), 966e968. Wijdicks, E. F., Bamlet, W. R., Maramattom, B. V., Manno, E. M., & McClelland, R. L. (2005). Validation of a new coma scale: The FOUR score. Annals of Neurology, 58(4), 585e593.

Please cite this article as: Bray, K et al., Procedural Pain in the Adult Neurological Intensive Care Unit: A Retrospective Study Examining Arterial Line Insertion, Pain Management Nursing, https://doi.org/10.1016/j.pmn.2019.09.003