Multidisciplinary Temperature Management in Patients with Hemorrhagic and Ischemic Stroke Cecelia Ratay RN BSN CCRN CNRN Senior Professional Staff Nurse, Neurovascular ICU UPMC Presbyterian 11 th Annual Acute Stroke Management Conference April 22, 2016
Background of the Problem o o o No protocol existed for management of a patient s fever, or temperature control complications such as shivering. Quality Improvement (QI) project completed in conjunction with the University of Pittsburgh School of Nursing QI Committee Members o Marilyn Hravnak PhD RN ACNP-BC FCCM FAAN o Jane Guttendorf DNP RN ACNP-BC CRNP o Lori Shutter MD o Kyra Klein MSN ACNP-BC CRNP o Benjamin Morrow MSN RN CCRN o Also reviewed by Joseph Durkin, PharmD
Background o Body temperature is regulated by feedback mechanisms in temperature sensitive, insensitive, and effector preoptic-anterior neurons of hypothalamus. Bader, 2013 o For each 1 C body temp, cerebral metabolic demands by 6-8%. Mrozek et al, 2012 o Meta-analysis of fever impact in patients with stroke and neurologic injury (n=14,431) o fever was associated with worse outcome in 7 measures of clinical, functional, and economic outcomes. Greer et al, 2008 o 2013 AHA/ASA guidelines for acute ischemic stroke management recommend: o causes of fever 38 C be identified o hyperthermia should be treated. Jauch et al, AHA/ASA Guideline, 2013
Background o Fever in neurologically injured patients causes: o Elevated levels of excitatory amino acids (like glutamate and dopamine) o Increased ischemic depolarizations (neurons repeatedly fire and die) o Blood-brain barrier breakdown o Impaired function of enzymes o Reduced cytoskeletal stability o Therefore, controlling fever may be beneficial Greer et al, 2008 o Interventions to control fever can include antipyretics, sedatives, surface cooling, internal blood cooling with endovascular catheters, and even pharmacological paralysis
Background: Temperature Control o o Cohort study of 18 aneurysmal SAH patients with refractory fever, therapeutic normothermia reduced cerebral metabolic distress o decreased lactate/pyruvate ratio (LPR: 40±24 versus 32±9, P<0.01) measured through cerebral microdialysis o lowered intracranial pressure (ICP: 32±11 versus 28±12 mmhg, P<0.05) Oddo et al, 2009 Comparative cohort study of TBI patients concluded that induced normothermia is effective in reducing fever burden and may reduce secondary injury, since it was shown to reduce intracranial pressure Puccio et al, 2009 o Case-control study of the impact of induced normothermia following aneurysmal SAH o compared 80 SAH patients undergoing aggressive vs conventional fever control. o Aggressive fever control patients trended to lower rate poor outcome at 3 months (AFC 44% vs CFC 60%, p = 0.1) o Trend became statistically significantly at 12 months (AFC 21% vs CFC 46%, p = 0.03). Badjatia et al, 2010
Background: Shivering o Receptors sense decreased temperature, leading to increased firing of cold sensitive neurons signaling heat production effector neurons o Shivering results in increased systemic metabolism (Bedside Shivering Assessment Scale correlates with Hypermetabolic Index) causes the patient to rewarm o Shivering is associated with significant reduction in brain tissue oxygenation Bader, 2013 o Increased expenditure of energy, consumption of oxygen, and production of carbon dioxide Badajita et al, 2008 o Limited interventions for shivering in the NVICU prior to this project.
Purpose o The purpose of this project was to: o increase the knowledge of critical care nurses regarding the importance of early recognition and intervention of fever in the critically ill patient with ischemic and hemorrhagic stroke o use multidisciplinary collaboration to develop, implement, and evaluate an evidence-based protocol for critical care nurses to initiate and maintain fever management in this patient population.
Specific Aims 1.With multidisciplinary collaboration, develop an evidence-based protocol for critical care nurses to initiate and maintain fever management, including management of shivering and infection surveillance. 2. Assess nursing knowledge of complications of fever in the neurologically injured patient, as well as knowledge of shivering and its management before an educational intervention and immediately after, as well as during the post-implementation phase. 3. Implement the protocol. 4. Evaluate the impact of the protocol on cumulative fever burden (total time patients are 38 C), frequency of infection surveillance, use of tiered shivering interventions. In addition, patient outcomes of patient length of stay in the NVICU and hospital, NIHSS on admission and unit/hospital discharge, GCS on admission and unit/hospital discharge, mrs as available from outpatient clinic follow-up will be evaluated before and after protocol implementation.
National Institute of Health Stroke Scale (NIHSS)
Glasgow Coma Scale (GCS)
Modified Rankin Scale (mrs)
Cumulative fever burden in this patient is 2 hrs + 2 hrs + 3 hrs = 7 hrs. 39 38.5 38 37.5 37 36.5 36 35.5 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Explanation of cumulative fever burden
Methods
Methods: Site o Setting o 20 Bed Neurovascular ICU (NVICU), UPMC Presbyterian o Medical care of these patients was directed by the Critical Care Medicine team, as well as Neurology and Neurosurgery o Institutional project approval by: o UPMC Quality Improvement Committee o UPMC Evidence-Based Practice Committee
Methods: Nursing Sample o NVICU Staff o ~60 registered nurses o Supervised by o Unit Director, two Clinicians, and two Clinical Resource Specialists
Methods: Patient Sample o Patients o Six month pre-implementation and post-implementation phases o The NVICU admits approximately 60 stroke patients per month o About 20% of these patients are expected to become febrile Rincon et al 2014 o Eligible were all consecutive adult patients registering at least one temperature >37.5 o C (this was the treatment threshold in the postimplementation group) in the NVICU and diagnosed with hemorrhagic or ischemic stroke who survived their hospitalization and to outpatient clinic follow-up.
Methods: Development of the Protocol o Review of the literature o Protocol reviewed by committee members (CR, MH, JG, LS, SC, BM, KK) and NVICU PharmD (JD) o 9-round iterative review process
Methods: Education o Developed standardized nurse education and knowledge surveys o Conducted unit-based journal club (Greer et al 2008 Meta-analysis) o Staff in-servicing, individual and small group basis o Over two weeks, for 20 minutes during scheduled shifts o Administered knowledge surveys (before journal club, after in-service, and again following protocol implementation)
Methods: Protocol Implementation o Protocols at every work station o Bedside nurses responsible for initiating protocol o Support from management and project leader (CR)
Methods: Evaluation o Nursing: o de-identified pre- and post-implementation knowledge survey scores o Patients (de-identified): o Cumulative fever burden in minutes before and after protocol implementation o Documentation of shivering assessment and management o Frequency of infection surveillance o Unit and hospital LOS, admission and ICU/hospital discharge NIHSS and GCS, and mrs from follow-up clinic as available
Data analysis o Staff Knowledge Survey Scores o Percent change from baseline o Patient data o Chi-Square for categorical and t-test for continuous variables, with statistical significance considered p < 0.05. o IBM SPSS
RESULTS
Results: Characteristics of patients (n = 67) Pre-Protocol n = 33 Post-Protocol n = 34 p value Age (mean years ±SD) 56.2 ±9.9 57.9 ±10.3 0.477 Gender n (%) Female Male 21 (63.6%) 12 (36.4%) 16 (47.1%) 18 (52.9%) 0.172 Diagnosis n (%) Subarachnoid Hemorrhage Cerebrovascular Accident Intracranial Hemorrhage 16 (48.5%) 9 (27.3%) 8 (24.2%) 13 (38.2%) 15 (44.1%) 6 (17.6%) 0.353 Admission NIHSS 12.8 ± 8.9 13.3 ± 8.6 0.814 Admission GCS 9.9 ± 3.9 10.9 ± 4 0.273
Results: Cumulative Fever Burden Cumulative Fever Burden in Minutes ALL (Mean ±Standard Deviation) Pre Post p value n = 33 2039.4 ± 1329.9 n = 34 1196.5 ± 1292.9 0.011 Diagnosis SAH n = 16 2451.3 ± 1464.3 n = 13 1779.23 ± 1756.9 CVA n = 9 1506.7 ± 785.4 n = 15 588 ±545 ICH n = 8 1815 ± 1405.6 n = 6 1455 ±848.5
Results: Outcomes of Patients Length of Stay (Days) ICU Hospital NIHSS At ICU Discharge Pre-Protocol Post-Protocol p value n =33 15.6 ±8.4 20.5 ±13.2 n = 33 11.8 ±8.2 n = 34 13.5 ±6.3 18.4 ±7.9 n = 31 10.9 ±8.7 0.263 0.437 0.963 At Hospital Discharge n = 30 9.4 ±7.1 n = 26 12.2 ±9.9 0.255 GCS At ICU Discharge n = 33 12.7 ±2.3 n = 34 12.5 ±2.5 0.706 At Hospital Discharge n = 31 13.1 ±1.9 n = 19 11.8 ±2.9 0.083 Modified Rankin Scale (1 to 4 months post injury)* n = 19 3.7 ±1.6 n = 21 3.4 ± 0.312 0.526
Results: Interventions to manage shivering Intervention Tier 1 Acetaminophen Buspirone Magnesium sulfate Warming blanket Tier 2 (Fentanyl or Dexmedetomidine) Tier 3 (both Fentanyl and Dexmedetomidine) Tier 4 (Propofoland Neuromuscular Blockade in an intubated patient) Post-protocol (n=34) n (%) 34 (100%) 9 (26.4%) 8 (23.5%) 11 (32.4%) 5 (14.7%) 0 (0%) 0 (0%)
Results: Patient infection surveillance Blood Cultures > 48h Pre-Protocol Post-Protocol p value n = 26 1.3 ±0.9 n = 30 1.1 ±1 0.421 Blood Cultures < 48h n = 7 2.9 ±0.9 n = 4 3.3 ±1.5 0.594 Urinalysis > 48h n = 29 1.4 ±1 n =29 1.1 ±0.7 0.186 Urinalysis < 48h n = 4 3 ±0.8 n = 5 3.8 ±1.3 0.323
Results: Staff Knowledge Score Pre n = 56 Post n = 50 3 months n = 43 Average score 70.7% 96.2% 80.5% Percentage of change from baseline 36.1% 13.9%
Clinical Implications o Use of an evidence-based, multidisciplinary protocol resulted in: o Statistically significant reduction in cumulative fever burden measured in minutes 38 C o Clinically significant reduction in ICU and hospital LOS by approximately 2 days
Limitations o Sample size may have been too small to detect meaningful changes in the GCS, NIHSS, or mrs scores, or we chose too short of an interval for follow-up assessment. o Since this was conducted as a quality improvement project and not as a research protocol with more stringent measurement validity and reliability, we experienced missing data points as well as retrospective score assignments. o Metrics may have been too blunt to assess and detect change in neurologic status and function. o May have been remiss to exclude patients who died in-hospital from the analysis. o May have contaminated our LOS assessments o Eliminating patients with neurologic deterioration then leading to death from the assessment also eliminated patients with poor scores
Conclusion o Demonstrated that a nurse-driven, evidence-based multidisciplinary protocol to apply fever and shivering management in a systematic and stepwise fashion was able to significantly reduce the time patients spent in a hyperthermic state o Trended towards shorter ICU and hospital lengths of stay by approximately two days. o A larger sample size, longer interval of follow-up, use of more sensitive outcome measures, and not excluding patients who died from the analysis would have been helpful in determining the protocol s impact on patient outcomes.
Acknowledgements o Thanks are extended to the Neurovascular ICU Staff, Katherine Spiering RN, MSN, Betsy George RN, PhD, and Joseph Durkin PharmD for contributing their expertise and support of this project. o My patient s father, referencing this Wall Street Journal article from November 23, 2005: http://online.wsj.com/ad/article/cigna/sb113271363926104904.html
References Bader MK. (2013 November 15). All shook-up: Managing shivering in therapeutic in therapeutic hypothermia [Webinar]. American Association of Critical-Care Nurses Webinar Series Badjatia N, Strongilis E, Gordon E, et al. Metabolic impact of shivering during therapeutictemperature modulation: The bedside shivering assessment scale. Stroke. 2008; 29:3242-3247. doi: 10.1161/STROKEAHA.108.523654 Badjatia N, Fernandez L, Schmidt JM, et al. Impact of induced normothermia on outcome after subarachnoid hemorrhage: A case-control study. Neurosurgery. 2010:66(4)606-701. doi: 10.1227/01.NEU.0000367618.42794.AA Choi H, Ko S, Presciutti M, et al. Prevention of shivering during therapeutic temperature modulation: The Columbia anti-shivering protocol. Neurocritical Care. 2011;14:389-394. doi:10.1007/s12028-010-9474-7 Greer DM, Funk SE, Reaven NL, Ouzounelli M, Uman GC. Impact of fever on outcome in patients with stroke and neurologic injury: A comprehensive meta-analysis. Stroke. 2008; 39:3029-3035. doi: 10.1161/STROKEAHA.108.521583 Jauch EC, Saver JL, Adams HP, et al. Guidelines for the early doi: 10.1161/STR.0b013e318284056a Mrozek S, Vardon, F, Geeraerts T. Brain temperature: Physiology and pathophysiology after brain injury. Anesthesiology Research and Practice. 2012. doi: 10.1155/2012/989487 Oddo M, Frangos S, Maloney-Wilensky E, et al. Effects of shivering on brain tissue oxygenation during induced normothermia in patients with severe brain injury. Neurocritical Care. 2010; 16:10-16. doi: 10.1007/s12028-009-9280-2 Oddo M, Frangos S, Milby A, et al. Induced normothermia attenuates cerebral metabolic distress in patients with aneurysmal subarachnoid hemorrhage and refractory fever. Stroke. 2009; 40:1913-1916. doi:10.1161/strokeaha.108.534115 Presciutti M, Bader MK, Hepburn M. Shivering management during therapeutic temperature modulation: Nurses perspective. Critical Care Nurse. 2012; 32(1):33-41. doi: 10.4037/ccn2012189 Puccio AM, Fisher MR, Jankowitz BT, et al. Induced normothermia attenuates intracranial hypertension and reduces fever burden after severe traumatic brain injury. Neurocritical Care. 2009; 11(1): 82-87. doi: 10.1007/s12028-009-9213-0 Rincon F, Hunter K, Schorr C, Dellinger RP, Zanotti-Cavazzoni S. The epidemiology of spontaneous fever and hypothermia on admission of brain injury patients to intensive care units: a multicenter cohort study. Journal of Neurosurgery. 2014; 121:950-960. Images from: http://www.japi.org/november_2012/07_oa_edaravone_citicoline_comparative_study.html, http://www.ebmedicine.net/topics.php?paction=showtopic&topic_id=123, and