1. SATs and SBTs: of Guidelines & Implementation
CUSP4MVP-VAP
SATs and SBTs: of Guidelines & Implementation
E. Wesley Ely, MD, MPH
Professor of Medicine
Vanderbilt University, Nashville, TN
VA TN Valley Health Care System GRECC

2. Disclosures: ICU Physician Vanderbilt - Abbott, Hospira, Orion - NIH and VA U.S. Federal Funding - Author of PAD Guidelines of SCCM Chair of SCCM Delirium section for PAD - Co-Chair of SCCM ICU Liberation project to aid world- wide implementation

3. Pain
Agitation
Delirium
Barr J, et al. Crit Care Med. 2013;41:

4. Barr J et al, CCM 2013;41:

5. New guidelines emphasize individual symptom management
OLD
(2002)
Clinical Practice Guidelines for the sustained use of sedatives and analgesics in the critically ill adult
Jacobi, CCM 2002
New
(2013)
Clinical Practice Guidelines for the management of Pain, Agitation, and Delirium in adult patients in the Intensive Care Unit
Barr, CCM 2013

7. Take Home Message

8. 50-70% Cognitively Impaired © rustyrhodes via Flickr
Wolters Intensive Care Med 2013; 39: 376
Jackson AJRCCM 2010; 182: 183
Girard Crit Care Med 2010; 38: 1513

9. Functionally Impaired
60-80%
Functionally Impaired
Marcel Oosterwijk via Flickr
Latronico Lancet Neurol 2011; 10: 931

10. ICU Survivorship Family Hobbies Work
Iwashyna Annals of Int Med 2010; 153:204-5

11. “
...like it was in a huge, empty gray space, sort of like a monstrous underground parking garage with no cars, only me, floating or seeming to float, on something…
-SB
“The time I spent seems like it was in a huge, empty gray space, sort of like a monstrous underground parking garage with no cars, only me, floating or seeming to float, on something. Every once in a while I would get to an edge of something horrible and once I remember I thought, "if I just let go, then this horror will be over.””
© Travis Smith via Flickr

12. Cognitive Impairment: Sepsis25
Before Sepsis
After Sepsis
p<0.001 20
Mild Cognitive Impairment
Moderate/Severe Cog Impairment 15
% survivors
cognitively impaired 10
Prospective cohort study of patients enrolled in the health and retirement study, an ongoing cohort of 27,000 community dwelling americans >50 years old. Every two years patients are assessed with good follow-up rates north of 90%. The GRAND IDEA put forth by this paper of ours was that there is a tripling of risk post sepsis (the main admission criteria for ICU patients) of moderate to severe cognitive impairment. That is, a HUGE increase in Demential like illness post ICU, and if delirium is the main risk factor for this, then modifying delirium could have a huge impact on a rising population of demented Americans.
1520 episodes of sepsis in 1194 patients over the 7 year study period
84% had normal cognition at baseline; 7.8% were mildly imparied and 8.7% were moderate/severely impaired
This modeling used in this study used within-person variation over time in cognitive function to estimate the impact of severe sepsis and to control for characteristics that did not change over time--in effect the patients served as their on controls over time. Thus, in the LR analysis, patients who developed severe sepsis were at 3.5 ( ) times the risk of developing moderate to severe cognitive impariment vs. those non-septic patients.
Severe sepsis was highly associated with progression to moderate/severe CI (OR 3.55; CI )
Patients NOT mechanically ventilated had a 4.4x great odds for developing moderate to severe CI after SS (CI ) 5
-3 years
-1 year
+1 year
+ 3 years
Iwashyna T, JAMA 2010;304:

13. Ely EW, JAMA 2004;291:

14. Delirium Duration & Mortality4
0 vs 1
HR 1.7
<0.001
0 vs 2
HR 2.69
0 vs 3
HR 3.73
p<.001 3
Relative Hazard of Death 2 1
NOTES: Crossing CI of HR from Cox regression happens based on a mathematical formula of Cox regression. Cox regression does not have an intercept in its equation, and lacking intercept does not provide a reference category when CI is shown graphically. For this reason, when a graph is drown, we need to pick an arbitral reference category where CI crosses on X axis, and typical default for this is to use median value of X. For example, HR is set to 1 at the median value (eg delirium duration =2 days), and HR shows relative increase or decrease from HR=1 (delirium duration = 2 days) when showing effect at non-median values.
Table 3. Multivariable analysis of cumulative number of delirium days
Number of Delirium
Days (p .001,
Overall)
30–Day Mortality
(p .02, Nonlinear Effect)a
Remaining Intubated
(p .001, Overall)
Remaining in Intensive Care Unit
(p .02, Overall)
HR 95% CI p HR 95% CI p HR 95% CI p
0 vs. 1 day – – –
0 vs. 2 days – – –
0 vs. 3 days – – –
HR, hazard ratio; CI, confidence interval. 1 2 3 4 5 6
Days of Delirium
Shehabi Y, et al. CCM 2010; 38:2311–2318

16. NEJM 2013;369:
Editorial by M. Herridge

17. Delirium and Brain Atrophy
Figure 2. Representative example of lateral ventricle size in 46-yr-old female and 42-yr-old female intensive care unit (ICU) survivors with no preexisting cognitive impairment: Axial T1-weighted brain images in two ICU survivors. A, Relatively normal ventricular volume (see arrow) in a 46-yr-old female who did not experience delirium in the ICU. Patient had a history of respiratory and heart failure. She was admitted to a medical ICU due to acute respiratory distress syndrome and was subsequently intubated and managed through the ICU without ever developing delirium. B, Enlarged ventricles (see arrow) in a 42-yr-old female who did develop delirium in the ICU. Patient was admitted to the hospital after reporting fever and dyspnea with a chest radiograph and other laboratory data confirming community acquired pneumonia and acute respiratory distress syndrome. The patient was admitted to the ICU and mechanically ventilated, experiencing 12 days of delirium and then resolution. There was no preexisting history of neurological impairment, and surrogate questioning for preexisting cognitive impairment was also negative.
(A) 46 year old, no delirium (B) 42 year old, 12 days of delirium
Gunther M et al. CCM 2012;40:

18. The Picture of Dementia Following ICU Care

19. Global Cognitive Scores by Age

20. Global Cognitive Scores by Age and Comorbidity

21. Primum non Nocere. - Hippocratic Oath
Primum non Nocere - Hippocratic Oath - First do no harm “Nothing to Fear but Fear Itself” - FDR inauguration, Overcome Fear of ICU Culture Change

22. So let’s focus on potentially modifiable aspects of care such as potent medications, delirium, and improving care and clinical outcomes…

23. Building blocks of managing Pain, Agitation & Delirium
ABCDEs:
Building blocks of managing
Pain, Agitation & Delirium E D C B A

24. Awake and Breathing Coordination
Duration of mechanical ventilation
Duration of coma
Mortality
Choose light sedation & avoid benzos
Duration of mechanical ventilation
Mortality
Delirium
Delirium monitoring & management
 Delirium detection
Early Mobility & Environment
Duration of delirium
Disability
ICU Length of Stay
Rehospitalization/Mortality
Morandi et al Curr Opin Crit Care 2011;17:43-9
Vasilevskis et al Crit Care Med 2010;38:S683-91
Vasilevskis et al Chest 2010;138:
Zaal et al, ICM 2013;39:481-88
Colombo et al, Minerva Anest 1012;78:

25. Pain, Agitation, and Delirium Are Interrelated
Barr J, et al. Crit Care Med. 2013;41:

26. “Pain should be routinely monitored in all adult ICU patients”
2013 PAD Guidelines:
“Pain should be routinely monitored in all adult ICU patients”
Grade 1B Recommendation
Crit Care Med. 2013;41:

27. Pain, Agitation, and Delirium Are Interrelated
Barr J, et al. Crit Care Med. 2013;41:

28. Targeted Level of Consciousness
Choose Target RASS
Assess Actual RASS
Modify treatment so Actual = Target

29. Grade 1B Recommendation
2013 PAD Guidelines:
“We recommend either daily sedation interruption or a light level of target sedation be routinely used…”
Grade 1B Recommendation
Crit Care Med. 2013;41:

30. Grade 1B Recommendation
2013 PAD Guidelines:
“We recommend that sedative medications be titrated to maintain a light* rather than deep level of sedation”
Grade 1B Recommendation
*Light sedation = RASS 0 to -2
Crit Care Med. 2013;41:

31. Awake and Breathing Coordination
Duration of mechanical ventilation
Duration of coma
Mortality
Choose light sedation & avoid benzos
Duration of mechanical ventilation
Mortality
Delirium
Delirium monitoring & management
 Delirium detection
Early Mobility & Environment
Duration of delirium
Disability
ICU Length of Stay
Rehospitalization/Mortality
Morandi et al Curr Opin Crit Care 2011;17:43-9
Vasilevskis et al Crit Care Med 2010;38:S683-91
Vasilevskis et al Chest 2010;138:
Zaal et al, ICM 2013;39:481-88
Colombo et al, Minerva Anest 1012;78:

32. 2 days Liberating from Ventilator SBT reduced weaning time by = 100 80
p<.001 60
Patients on Ventilator (%) 40
Control (n =151) 20
Protocol (n =149) 30 20 10 15 5 25
Time (Days)
Ely EW, et al. N Engl J Med 1996;335:1864-9

33. 2 days Liberating from Sedation SAT reduced ventilator time by = 10080 60
Patients on Ventilator (%) 40
Control (n=60)
Adjusted p<.001 20
Protocol (n=68) 5 10 15 20 25 30
Time (Days)
Kress JP, et al. N Engl J Med 2000;342:1471-7

34. SAT + SBT = 4 day shorter ICU/hosp LOS

35. ABC Trial: One-Year Survival
100
NNT=7 80
ABC approach (n=167) 60
Patients Alive (%) 40
Control (n=168) 20
p=.01 60
120
180
240
300
360
Days
Girard TD, et al. Lancet 2008;371:126-34

36. Sedation Interruption in SLEAP
Mehta S, JAMA 2012;308:

37. Benzodiazepine Use in Trials *
Study
Control
Treatment
Kress NEJM 2000
90 mg/day
53 mg/day
Girard ABC Lancet 2007
84 mg/day
54 mg/day
Mehta SLEAP JAMA 2012
82 mg/day
102 mg/day
OSCILLATE NEJM 2013
141 mg/day
199 mg/day
* All values converted and expressed as mean midazolam dose per patient, median for ABC study were 8 mg and 5 mg, respectively

38. SPICE Study – first 48 hours mean 50 mg/d benzos
Pratik wrote: I just read the Shehabi SPICE study (AJRCCM 2012) showing early deep sedation is associated with mortality (apart from delayed extubation) in detail and recommend it as reading for those who have not. Its got some neat data about sedation practices (albeit in Australia/NZ) that still supports low implementation of targeted sedation and of SATs (20% in Australia). There are also some nice descriptive data about prevalence of delirium over time.
In this study early deep sedation (first 48 hours) was not associated with delirium. Some of you may be asked about this on the podium one day so I thought I’d share a few thoughts. Also these delirium results are different from the diurnal sedation paper we just published in CCM (Seymour et al.) but they are answering different questions tham we did.
1.       They defined deep sedation of RASS -3 to -5 so its not just coma
2.       Delirium was measured only in patients who were RASS -2 to +1. So some cases Im sure got missed at -3 and then in the more positive range of the RASS
3.       All sedatives were lumped together in the delirium analysis since they chose the risk factor of deep sedation irrespective of how you got there. They did separate out the sedatives for the liberation from MV outcome and found that midazolam but not propofol was associated with the outcome.
4.       There may have been some missing delirium assessments even in the RASS -2 to +1 range- perhaps those were the aberrant UTAs that show up which almost always tend to be CAM positive. They clearly mention the number of missing values on day 1 on the study but not during the outcome phase.
Pratik
Shehabi AJRCCM 2012;186:724-31

39. Awake and Breathing Coordination
Duration of mechanical ventilation
Duration of coma
Mortality
Choose light sedation & avoid benzos
Duration of mechanical ventilation
Mortality
Delirium
Delirium monitoring & management
 Delirium detection
Early Mobility & Environment
Duration of delirium
Disability
ICU Length of Stay
Rehospitalization/Mortality
Morandi et al Curr Opin Crit Care 2011;17:43-9
Vasilevskis et al Crit Care Med 2010;38:S683-91
Vasilevskis et al Chest 2010;138:
Zaal et al, ICM 2013;39:481-88
Colombo et al, Minerva Anest 1012;78:

40. No Sedation: ICU Length of Stay
100 80
Control (n=58) 60
Patients Remaining in ICU (%) 40
Intervention (n=55) 20
9.7 days
ICU stay reduced by 7 14 21 28
Days
Strom T, et al. Lancet 2010;375:475-80

41. Grade 2B Recommendation
2013 PAD Guidelines:
“We suggest that sedation strategies using non-benzodiazepines (propofol or dexmedetomidine) may be preferred over sedation with benzodiazepines (midazolam or lorazepam)”
Grade 2B Recommendation
Crit Care Med. 2013;41:

42. Pain, Agitation, and Delirium Are Interrelated
Barr J, et al. Crit Care Med. 2013;41:

43. Awake and Breathing Coordination
Duration of mechanical ventilation
Duration of coma
Mortality
Choose light sedation & avoid benzos
Duration of mechanical ventilation
Mortality
Delirium
Delirium monitoring & management
 Delirium detection
Early Mobility & Environment
Duration of delirium
Disability
ICU Length of Stay
Rehospitalization/Mortality
Morandi et al Curr Opin Crit Care 2011;17:43-9
Vasilevskis et al Crit Care Med 2010;38:S683-91
Vasilevskis et al Chest 2010;138:
Zaal et al, ICM 2013;39:481-88
Colombo et al, Minerva Anest 1012;78:

44. Cardinal Symptoms of Delirium and Coma
Morandi A, et al. Intensive Care Med. 2008;34:

45. “We recommend routine monitoring for delirium in adult ICU patients”
2013 PAD Guidelines:
“We recommend routine monitoring for delirium in adult ICU patients”
Grade 1B Recommendation
Crit Care Med. 2013;41:

46. Delirium and Executive Function

47. If delirium is not screened for using a validated delirium screening tool it is missed ~75% of time.
Inouye SK Arch Intern Med. 2001;161:
Devlin JW Crit Care Med. 2007;35:
Spronk PE Intensive Care Med. 2009;35:
van Eijk MM Crit Care Med. 2009;37:

48. Ely EW, JAMA 2001;286:

49. Ely EW, JAMA 2003;289:
The Miraculous Haul of Fishes, by Henry Tanner, who was the first African American to become a member of the National Academy of Design. This from 1927 confirmed his reputation as the foremost biblical painter of his day. Probably from Luke 5:1-11 (maybe from John’s story, but details make Luke more likely).

50. Don’t forget about Dr. DRE
Diseases
Sepsis, COPD, CHF
Drug Removal
SATs and stopping benzodiazepines/ narcotics
Environment
Immobilization, sleep and day/night, hearing aids, glasses, noise
Medical Intensive Care Unit

51. Brain Road Map (A framework for bedside rounds)
1. Where is the patient going?
Target RASS
2. Where is the patient now?
Current RASS
Current CAM-ICU
© Brian Sloan via Flickr
3. How did they get there?
Drugs

52. Jiro Dreams of Sushi - Tokyo
Excellence
Aristotle: “We are what we repeatedly do Excellence is not an act, but a habit”
Jiro Dreams of Sushi - Tokyo

53. Building blocks of managing Pain, Agitation & Delirium
ABCDEs:
Building blocks of managing
Pain, Agitation & Delirium E D C B A

54. “ “
I survived and that is the main thing. And I am so grateful to God that I survived and am now off all oxygen and consider myself all well except that I can’t remember to take my medications...
-SB
© Cappi Thompson via Flickr

55. The ICU Delirium and Cognitive Impairment Study Group at the Loveless Café, Nashville TN

56. ICU Delirium and Cognitive Impairment Study Group: selected local members
Pratik Pandharipande
Jim Jackson
Jin Han
Ed Vasilevskis
Chris Hughes
Alessandro Morandi
Paula Watson
Lorraine Ware
Gordon Bernard
Bob Dittus
Ted Speroff
Wes Ely
Leanne Boehm
Joyce Okahashi
Cayce Strength
Brenda Pun
Lauren Hardy
Amy Lipsey
Ryan Black
Jessica McCurley
Michael Santoro
Carrie Jones
Morgan Crawford
Mayur Patel
Tim Girard
John Gore
Baxter Rogers
Stephan Heckers
Cathy Fuchs
Heidi Smith
Ty Berutti
Brad Strohler
Elizabeth Card
Jennifer Thompson
Ayumi Shintani
Stephanie Hamilton

58. Key Epidemiological Points:
Patients suffer from long-lasting and disabling aspects of critical illness that demand our attention as a medical community
Acquired or accelerated cognitive impairment is a major public health problem following ICU care for both the old and young
This cognitive impairment appears most pronounced in domains of executive dysfunction and memory
Frontal lobe and hippocampal atrophy are being consistently found in recent studies
This injury is likely distinct from or complementary to Alzheimer’s pathology, though we are in our infancy in learning about this entity (e.g., large pathology study under review)
Delirium and drug exposure appear to be the most modifiable aspects of care, with need for more trials to delineate next steps

59. Key Management Points:
Establish an overarching protocolized approach to daily ICU patient management using 2013 PAD Guidelines
Assess & treat pain first (may be sufficient)
If patient remains agitated after adequately treating pain, use prn/bolus sedation initially, if frequent boluses (>3/hr) use continuous sedation
Avoid benzodiazepines in most patients
Turn off sedation daily and restart only if needed at lowest dose to maintain chosen target level of consciousness
Deep sedation (RASS -4/-5) appears harmful; target awake/alert
Screen for delirium (CAM-ICU or ICDSC); If delirious, first seek reversible causes and attempt non-pharmacologic management
Use the ABCDEs to improve outcomes for your patients

60. CUSP4MVP-VAP project measures
As part of this project, teams will collect and receive reports for metrics to support your improvement efforts.
We held calls with your data facilitator. The data collection tool, including instructions are available at:
Medical Intensive Care Unit

61. CUSP4MVP Data Collection Sedation and Delirium
1) Percentage of RASS/SAS actual being {-1, 0, 1} or {4, 5}
2) Percentage of achieving RASS/SAS target
Distribution of RASS/SAS actual scores
4) Delirium assessment compliance rate
5) Percentage of incorrectly reporting CAM-ICU/ ASE UTA (higher is worse)
6) Percentage of CAM-ICU negative or ASE <=2 (no delirium)
Medical Intensive Care Unit

62. CUSP4MVP Data Collection SAT/SBT (next call on March 18)
(1) SAT compliance rate (2) SBT compliance rate (3) SAT contraindication rate (4) SBT contraindication rate (5) Percentage of ventilated patient days without sedation (6) SBT with Seds off compliance rate (7) SAT contraindication distribution plot and table (counts and percentages) (8) SBT contraindication distribution plot and table (counts and percentages)
Medical Intensive Care Unit

63. Your Next Steps:
Share your protocols regarding sedation and delirium management; will share with other participating teams to
Review data collection requirements and develop plan to collect and submit data.
Medical Intensive Care Unit

64. Reminder: By Content Call, Module 5 (April 1, 2014):
Watch the Science of Safety (SOS) Video
tools/cusptoolkit/videos/04a_scisafety/index.html
Develop a method to deliver the SOS Video to your entire unit’s staff
Administer the SSA and submit aggregated results to CE
Facilitate at least one team meeting

65. Next Call
March 18, 2014
SAT / SBT
Dr Mike Klompas