Portal de Periódicos da CAPES

End of Life in the Neurological Intensive Care Unit: Is Extubating to Comfort Care Comfortable?

2019; Elsevier BV; Volume: 58; Issue: 4 Linguagem: Inglês

10.1016/j.jpainsymman.2019.07.004

1873-6513

Angela Yeh, Christopher Pietras, David Wallenstein, Peifeng Hu,

Intensive Care Unit Cognitive Disorders

Patients extubated to comfort care in the neurological intensive care unit (Neuro-ICU) comprise a unique population because their end-of-life signs of distress can be related to both neurological pathology and multisystem dysfunction. Previous studies have examined time to cardiac death after extubation to comfort care and its predictors,1Huynh T.N. Walling A.M. Le T.X. et al.Factors associated with palliative withdrawal of mechanical ventilation and time to death after withdrawal.J Palliat Med. 2013; 16: 1368-1374Crossref PubMed Scopus (30) Google Scholar, 2Varelas P.N. Hacein-Bey L. Schultz L. et al.Withdrawal of life support in critically ill neurosurgical patients and in-hospital death after discharge from the neurosurgical intensive care unit.J Neurosurg. 2009; 111: 396-404Crossref PubMed Scopus (13) Google Scholar but there are few studies describing patients' distress developed after extubation and the time required to achieve observable distress control.3Kompanje E.J.O. van der Hoven B. Bakker J. Anticipation of distress after discontinuation of mechanical ventilation in the ICU at the end of life.Intensive Care Med. 2008; 34: 1593-1599Crossref PubMed Scopus (73) Google Scholar, 4Blinderman C.D. Billings A. Comfort care for patients dying in the hospital.N Engl J Med. 2015; 373: 2549-2561Crossref PubMed Scopus (88) Google Scholar, 5Von Gunten C. Weissman D.E. Fast Fact and Concept #033 and #034: Ventilator Withdrawal Protocol.2001https://www.eperc.mcw.eduGoogle Scholar, 6Truog R.D. Campbell M.L. Curtis J.R. et al.Recommendations for end-of-life care in the intensive care unit: a consensus statement by the American College of Critical Care Medicine.Crit Care Med. 2008; 36: 953-963Crossref PubMed Scopus (667) Google Scholar, 7Rubenfeld G.D. Crawford S.W. Principles and practice of withdrawing life-sustaining treatment in the ICU. Oxford University Press, New York2001: 127-147Google Scholar The goal of this study is to determine if Neuro-ICU patients demonstrate persistent signs of distress after transitioning to comfort care and how long these take to treat. This was a retrospective medical record review of all patients who died in the Neuro-ICU at one tertiary academic hospital from October, 2016, to October, 2017. The study was exempt from institutional review board oversight because it involved only patients who had died before the investigation. Inclusion criteria were neurological injury as cause of death and mechanical ventilation with extubation to comfort care. Exclusion criteria included age less than 18 years, pregnancy, penetrating brain injury, brain death, and receiving cardiopulmonary resuscitation within 24 hours of cardiac death. Also excluded were deaths with incomplete documentation of life support withdrawal, medication administration, and responses to distress. Neuro-ICU patients whose physical location of death was in another unit were also excluded because of logistical difficulties. The study's primary outcomes were to determine if neurologically devastated patients developed distress after extubation to comfort care and whether this persisted. Secondary outcomes included both time to distress control and time to cardiac death. Because the patients evaluated during this study were unconscious, signs of distress were collected from nursing notes, the documented reason for as needed (PRN) medication administration, and from the Pain Assessment Tool for Non-Cognizant Adults (Non-Cog), developed by University of California, Los Angeles, and used primarily in the intensive care units to evaluate for pain in nonverbal or unconscious adults.8Siegel C. Pain Assessment Tool for Non-Cognizant Adults.https://www.uclahealth.org/palliative-care/Workfiles/Pain-Assessment-Tool-Non-Cognizant.pdfGoogle Scholar Because these patients were neurologically devastated and unable to report symptoms, the term “signs of distress” was used rather than “symptoms.” Developing signs of distress immediately after extubation was defined as within 30 minutes after extubation. A persistent sign of distress was defined as requiring greater than 50% of the time between extubation and cardiac death to achieve control. We defined distress control as not receiving any PRN medications nor requiring any opioid dose adjustments to infusions or boluses for at least 1 hour. Time to distress control was reported in hours and derived from objective scales such as the Non-Cog, subjective nursing documentation, and the medication administration record. Continuous data such as time required to achieve distress control and time from extubation to cardiac death were reported as means, medians, and standard deviations (SD). Of the 79 deaths in the Neuro-ICU whose charts were reviewed, 41 were excluded (16 did not meet inclusion criteria, 22 met exclusion criteria, two had poor documentation, and one was restricted). The mean age of the 38 patients whose records were selected for review was 63.1 years (median 65.0, SD 16.9, range 20–96.9), with 22 patients (58%) identified as female. The three most common diagnoses before Neuro-ICU admission included hypertension (44.7%), malignancy (31.6%), and atrial fibrillation (21.1%). The most prevalent diagnoses for admission to the Neuro-ICU were nontraumatic subarachnoid or intracerebral hemorrhage (28.9%), traumatic subarachnoid or subdural hemorrhage (21.1%), cerebral infarction (21.1%), and neurological malignant neoplasm—primary or secondary (13.2%). Of the 38 patients, 28 (73.7%) developed signs of distress, with 23 (60.5%) patients developing signs of distress immediately after extubation. Table 1 displays initial signs of distress observed or documented by the nurse. Twenty (52.6%) patients developed signs of distress that took >50% of their remaining time after extubation to attempt to achieve distress control. The median time in attempting to achieve distress control was 2.4 hours (mean: 4.1 hour, SD: 4.7 hours, range: 0.2–18.8 hours). The median time to cardiac death was 2.7 hours (mean: 4.7 hours, SD: 5.8 hours, range: 0.1–20.5 hours).Table 1Initial Sign of Distress Developed After Extubation of Patients to Comfort Care in the Neurological Intensive Care Unit (n = 28)Sign of DistressNRespiratory symptom (e.g., labored breathing, increased work of breathing, dyspnea, coughing, or terminal congestion)9Pain (intermittent pain expression or perceived pain)3Anxiety, agitation, delirium1Vital sign abnormality (e.g., heart rate, blood pressure, or respiration rate ≥ 10% above baseline)1Signs of distress (e.g., dilated pupils, perspiring, flushing, diaphoretic, abnormal movements)0Pain + vital sign abnormality + signs of distress1Respiratory symptom + pain6Respiratory symptom + signs of distress1Respiratory symptom + vital sign abnormality4Respiratory symptom + pain + vital sign abnormality1Respiratory symptom + pain + vital sign abnormality + signs of distress1 Open table in a new tab All patients who were extubated to comfort care were premedicated with an opioid. Twenty-seven patients received a morphine infusion initiated at a median time of 0.8 hours before extubation (mean: 3.3 hours, SD 6.3 hours, range: 0.2–24 hours). Twenty-two patients were premedicated with an anticholinergic. Twelve of those patients received transdermal scopolamine placed at a median time of 0.7 hours before extubation (mean 3.6 hours, SD 6.7 hours, range: 0.03 hours–23.4 hours). Four of the 38 patients (10.5%) received a benzodiazepine as premedication before extubation. Of the 28 patients who developed signs of distress, the median intravenous morphine equivalent dosage received from time of extubation to time of death was 39.1 mg (mean: 58.5 mg, SD 56.4 mg, range: 2–229.9 mg). Nine of these patients also received lorazepam. The median dose of lorazepam received was 1 mg (mean: 3.22 mg, SD 5.7 mg, range: 0.5–18 mg). Little is known about the frequency of signs of distress after extubation to comfort care, how fast these present themselves, and how quickly they are controlled. This study is unique in demonstrating that 74% of patients in the Neuro-ICU who were extubated to comfort care developed some sign of distress after extubation. Of the patients who developed signs of distress, 82% developed a sign of distress in the first 30 minutes after extubation, and 71% required more than 50% of their remaining time after extubation until cardiac death to achieve distress control. These retrospective data results suggest that not only do the majority of patients who are extubated to comfort care and remain in the Neuro-ICU develop signs of distress but that these may develop shortly after extubation and persist for much of the time until cardiac death. The possibility of inadequate premedication is considered in the 23 patients who developed signs of distress immediately after extubation. Because circulating plasma levels of transdermal scopolamine are detected within 4 hours after placement and peak levels are usually obtained within 24 hours,9Novartis Consumer Health, IncTransderm Scop (Scopolamine) transdermal system patch.2013https://www.accessdata.fda.gov/drugsatfda_docs/label/2013/017874s038lbl.pdfGoogle Scholar ineffective premedication could contribute to comfort care patients' signs of respiratory distress or distress related to terminal congestion. Although all patients had a neurological pathology contributing cause of death, benzodiazepines were infrequently used before and after extubation. While national standard guidelines have not been set for extubation to comfort care, most articles, studies, and protocols recommend prophylactic intravenous bolus dose of both an opioid and a benzodiazepine.2Varelas P.N. Hacein-Bey L. Schultz L. et al.Withdrawal of life support in critically ill neurosurgical patients and in-hospital death after discharge from the neurosurgical intensive care unit.J Neurosurg. 2009; 111: 396-404Crossref PubMed Scopus (13) Google Scholar, 3Kompanje E.J.O. van der Hoven B. Bakker J. Anticipation of distress after discontinuation of mechanical ventilation in the ICU at the end of life.Intensive Care Med. 2008; 34: 1593-1599Crossref PubMed Scopus (73) Google Scholar, 4Blinderman C.D. Billings A. Comfort care for patients dying in the hospital.N Engl J Med. 2015; 373: 2549-2561Crossref PubMed Scopus (88) Google Scholar, 5Von Gunten C. Weissman D.E. Fast Fact and Concept #033 and #034: Ventilator Withdrawal Protocol.2001https://www.eperc.mcw.eduGoogle Scholar, 6Truog R.D. Campbell M.L. Curtis J.R. et al.Recommendations for end-of-life care in the intensive care unit: a consensus statement by the American College of Critical Care Medicine.Crit Care Med. 2008; 36: 953-963Crossref PubMed Scopus (667) Google Scholar, 7Rubenfeld G.D. Crawford S.W. Principles and practice of withdrawing life-sustaining treatment in the ICU. Oxford University Press, New York2001: 127-147Google Scholar Because patients extubated to comfort care in the Neuro-ICU are often unconscious and are unable to communicate, health care providers and family members are unable to know with certainty whether the observed signs of distress are resulted from physical discomfort or altered physiology related to either the injury or the dying process. Our data have shown that respiratory distress was either the primary sign of distress or a component of distress that initially developed; however, signs of respiratory distress may be manifestations of disorders of ventilator control due to neurological pathology. In these situations, providers may need to provide education to family members and consider palliative sedation before and during extubation to prevent the perception of respiratory distress, such as labored breathing, tachypnea, and terminal congestion. Limitations of this study include potential for sample bias related to the single site of analysis and the small sample size. Other limitations include the retrospective design and the risk of observer bias and documentation bias. The data also revealed that medications were generally provided for observations such as vital sign abnormalities or dilated pupils as signs of distress or symptomatology in patients with neurological pathology. Despite these limitations, to the authors' knowledge, this is the first study to examine signs of distress that develop after extubation to comfort care in neurologically devastated patients. The findings from this descriptive study could help educate not only palliative care physicians but also any physician who will be caring for patients extubated to comfort care from the Neuro-ICU on what they could possibly expect in the development of signs of distress. This could hopefully change how providers approach patients who are planned for extubation to comfort care in terms of premedication or post-extubation liberalization of medications. Almost 74% of patients extubated to comfort in the Neuro-ICU are likely to develop signs of distress requiring pharmacologic intervention; and of these patients who develop signs of distress, over 70% will develop persistent signs of distress that will be difficult to control.