Washington University Emergency Medicine Knowledge Translation Project

1
Washington University Emergency Medicine
Knowledge Translation Project

• Utilization of Corticosteroids in Bacterial
  Meningitis Bringing the Evidence to the Bedside

2
Objectives

• Knowledge Translation Review
• Wash U EM KT Project Methods
• Wash U EM KT Project Results
• KT gap identification
• Retrospective chart review
• Internal Survey
• External Survey

3
What is Knowledge Translation?

1. I have no idea
2. I dont care
3. Behavioral science
4. Translating foreign manuscripts
5. The science of moving from evidence to action
6. Being made aware of evidence through Journal Club
   or CME

4
What Is Knowledge Translation?

• A dynamic process that includes synthesis,
  dissemination, exchange, and ethically sound
  application of knowledge to provide more
  effective health care services and products to
  strengthen the health care system
• Canadian Institutes of Health Research
• A structured process that results in
  scientifically researched evidence being used to
  improve the outcomes of professional practice

5
Traditional Journal Club formats are Knowledge
Translation.

1. Yes
2. No
3. Uncertain

6
3 Distinguishing Features

• Emphasizes the process of qualifying research
  evidence
• Supports the flow of data and understanding from
  researcher to practitioner or policy-maker (as
  well as the flow from practitioner and
  policy-maker to researcher)
• Pertains to complex social and organizational
  contexts

7
Why Is Knowledge Translation Important?

• The creation of new knowledge often does not on
  its own lead to widespread implementation or
  impacts on health
• One study, published in the New England Journal
  of Medicine, looking at 12 hospitals, gt400
  quality indicators and 6,700 patients, found
  that 45 of patients do not receive recommended
  care
• A significant amount of time is needed for
  scientific research to be incorporated into
  clinical practice (6-13 years)

8
Steps of Knowledge Translation

• Synthesis -- the contextualization and
  integration of research findings of individual
  research studies within the larger body of
  knowledge on the topic
• Must be reproducible and transparent in its
  methods, using quantitative and/or qualitative
  methods
• Systematic reviews, methods developed by the
  Cochrane Collaboration, results of a consensus
  conference or expert panel
• Realist syntheses, narrative syntheses,
  meta-analyses, meta-syntheses and practice
  guidelines
• Dissemination -- identifying the appropriate
  audience, and tailoring the message and medium to
  the audience
• Educational sessions with patients, practitioners
  and/or policy makers
• Developing and executing the dissemination and
  implementation plan, tools creation, and media
  engagement

9
Steps of Knowledge Translation

• Exchange the interaction between the knowledge
  user and the researcher resulting in mutual
  learning
• Collaborative problem-solving between researchers
  and decision makers that happens through linkage
  and exchange, resulting in mutual learning
  through the process of planning, producing,
  disseminating, and applying existing or new
  research in decision-making
• Ethically sound -- must be consistent with
  ethical principles and norms, social values, as
  well as legal and other regulatory frameworks
• Keep in mind that principles, values and laws can
  compete among each other at any given point in
  time
• Application of knowledge -- the iterative process
  by which knowledge is put into practice

10
Knowledge Cycle
11
What Significant KT Gap was identified at BJH?

• The second-year class proposed that we were not
  following evidence-based guidelines regarding the
  utilization of adjunctive corticosteroids in the
  treatment of acute bacterial meningitis
• Other KT gaps considered
• Use of Plavix in ACS/STEMI
• Utility of LP after negative head CT in SAH
• Utilization of Canadian C-spine/Nexius criteria
  when ordering C-spine films
• Need for pre-oxygenation in procedural sedation
• A retrospective chart review was performed to
  assemble objective evidence of this gap in
  Knowledge Translation
• All ED patient charts within a 1 year period
  (9/1/06 to 8/30/07) were examined for inclusion
  into the study
• All patients given the diagnosis of meningitis or
  encephalitis who underwent lumbar puncture were
  included
• Patients meeting SIRS criteria were excluded from
  the study

12
How many bacterial meningitis patients have you
diagnosed over the last 12 months in the ED?

1. None
2. 1
3. 2
4. 3
5. 4
6. gt 5

13
I read all of tonights Journal Club articles.

1. Yes
2. No

14
In definite bacterial meningitis, medical
evidence suggests a role for steroids.

1. Yes
2. No

15
The articles distributed for tonight changed my
opinion about the use of steroids in meningitis.

1. Yes, from use to dont use
2. Yes, from dont use to use
3. No

16
In most patients I evaluate, I cannot clearly
distinguish bacterial meningitis from other
potential sources of symptoms before CSF results
are available.

1. Yes
2. No

17
Background on the Use of Steroids in Acute
Bacterial Meningitis (ABM)

• Despite advances in vaccinations and antibiotic
  therapy, significant morbidity and mortality
  exists in patients with ABM
• There are approximately 25,000 cases of ABM
  yearly
• Review of the literature demonstrates that
  administering corticosteroids with antibiotics
  improves outcomes (especially in cases of Strep.
  pneumoniae)
• Inflammatory responses within the CNS are thought
  to cause increased morbidity and mortality in
  these patients
• Administration of dexamethasone (10 mg IV Q 6
  hours) has shown clinical efficacy in reducing
  morbidity and mortality when given with (or prior
  to) appropriate antibiotic therapy

18
Inflammatory Cascade in ABM
19
(No Transcript)
20
(No Transcript)
21
(No Transcript)
22
(No Transcript)
23
Retrospective Review Data

• 65 patients had a diagnosis of meningitis or
  encephalitis
• 2 were excluded for repeat visits
• 4 were excluded because no LP was performed
• 15 of the remaining 59 patients met the gold
  standard criteria for the diagnosis of meningitis
  with a positive CSF culture
• 2 were positive for Strep. pneumoniae
• 1 was positive for Strep. dysgalactiae
• 1 was positive for coagulase-negative Staph.
• 7 were positive for HSV
• 1 was positive for enterovirus
• 1 was positive for arbovirus
• 2 were positive for Cryptococcus

24
Retrospective Review Data

• 8/59 patients received steroids (13.6)
• 4/8 received steroids before or at the time of
  antibiotic administration (6.8)
• 2/8 had positive CSF cultures (both with Strep.
  pneumoniae)
• Conclusion Steroids are not given routinely
  prior to antibiotic administration for presumed
  meningitis despite current recommendations

25
How Large is the KT Gap at BJH?

• A survey was distributed to Wash U Emergency
  Medicine residents and attendings asking
  questions about the use of steroids in ABM in
  their clinical practice
• 76 of 88 (86.4) responded
• 31 of 76 (40.8) indicated routinely using
  steroids in suspected cases of meningitis in
  adults
• 45 of 76 (59.2) do not routinely use steroids
• 35 of 74 (47.3) indicated using steroids in
  suspected cases of meningitis in pediatric
  patients
• 39 of 74 (52.7) do not use steroids in suspected
  cases of meningitis in pediatric patients

26
How Large is the KT Gap at BJH?

• 33/76 (43.4) indicated that they administer
  steroids prior to antibiotic therapy
• 19/76 (25.5) administer steroids with
  antibiotics
• 7/76 (9.2) administer steroids post-antibiotics
• 21/76 (27.6) do not use steroids at all
• In summary, less than half of the physicians
  surveyed routinely use steroids in cases of
  suspected ABM
• 88 of the physicians considering steroids as
  adjunctive therapy indicated that they would
  administer steroids prior to, or concurrent with,
  antibiotics

27
I almost always obtain a head CT before
performing an LP.

1. Yes
2. No

28
Surprise Finding!

• Pre-LP CT obtained in 100 of cases!

29
Literature Review
30
Knowledge Translation Pipeline
31
How Were Leaks in the Pipeline Identified at BJH?

• In order to identify specific leaks in the KT
  Pipeline, survey questions were designed to
  address hypothesized barriers in bringing the
  evidence to the bedside
• Responses were analyzed and categorized for each
  step in the KT Pipeline

32
Awareness

• Leaks in this leg of the Pipeline include
• Information overload
• Literature search deficiency
• Inadequate time
• Insufficient strategies
• Proposed solutions
• Accessible appraisals
• Automatic delivery of newsworthy, clinically
  relevant evidence

33
Awareness

• Are residents and attendings unaware of the large
  amount of research data supporting the use of
  corticosteroids as an effective adjunctive
  therapy for ABM?
• Yes 13/74 (17.6)
• No 61/74 (82.4)
• I have never been taught that steroids were an
  important therapy in ABM
• Yes 27/75 (36.0)
• No 48/75 (64.0)
• I have no idea what dose of steroids is
  appropriate
• Yes 34/76 (44.7)
• No 42/76 (55.3)

34
Awareness

• I have no idea when steroids should be
  administered in conjunction with antibiotics for
  ABM
• Yes 21/76 (27.6)
• No 55/76 (72.4)
• Resources used to keep up-to-date on the latest
  EM literature
• Wash U EM Journal Club (71.2)
• Wash U EM Lecture Series (54.5)
• EM Rap (30.3)
• EM Abstracts (25.8)
• EM Journal Watch (21.2)
• Audio Digest (19.7)
• Practical Reviews in EM (10.6)
• BEEM (1.5)
• Many others Uptodate, eMedicine, NEJM, Annals,
  ACEP News, LLSA, etc.

35
Acceptance

• Leaks in this leg of the Pipeline include
• Competing influences
• Marketing
• Authoritarian doctrine
• Contradictory experience
• Poorly-differentiated healthy vs. unhealthy
  skepticism
• Proposed solutions
• Tort reform
• Interdisciplinary education
• Balanced commercials

36
Acceptance

• Does current research data support the use of
  steroids in ABM?
• Yes 53/75 (70.7)
• No 29/75 (29.3)
• 4 respondents stated that steroids were
  beneficial depending on the organism involved
• 2 respondents stated that steroids may be
  beneficial in patients with a high suspicion of
  ABM
• 5 respondents stated that current literature only
  supports using steroids in pediatric patients
  with ABM
• 3 respondents stated that the literature was
  thin or conflicting, or that the impact is
  limited
• 2 respondents were unfamiliar or not clear on
  the literature

37
Acceptance

• My physician peers doubt the efficacy of steroids
  in ABM
• Yes 35/69 (50.7)
• No 34/69 (49.3)
• My nursing peers doubt the efficacy of steroids
  in ABM
• Yes 22/69 (31.9)
• No 47/69 (68.1)

38
Applicable

• Leaks in this leg of the Pipeline include
• Uncertain interpretation
• Uncertain local demographics
• Clinically significant differences from study
  population
• Proposed solutions
• Increased pragmatic clinical trials
• Clear description of demographics

39
Applicable

• Findings which would increase the likelihood to
  use steroids in suspected ABM
• Abnormal CSF white count (88.6)
• Any combination of altered mental status, stiff
  neck, photophobia (79.1)
• Positive CSF gram stain (77.6)
• Abnormal CSF glucose (56.7)
• Altered mental status (56.7)
• Fever (55.2)
• Petechial rash (55.2)
• Stiff neck (49.3)
• Abnormal CSF protein (47.8)
• Headache (34.3)
• Elevated LP opening pressure (31.3)

40
Able

• Leaks in this leg of the Pipeline include
• Resource constraints
• Skill competence
• Proposed solutions
• Original research description of evidence use at
  dissimilar clinical settings

41
Able

• ABM often presents atypically, limiting my
  ability to recognize it as the source of
  presenting symptoms in time to permit
  administration of steroids before antibiotics
• Yes 38/76 (50.0)
• No 38/76 (50.0)

42
Act On

• Leaks in this leg of the Pipeline include
• Hectic ED environment
• Frequent distractions
• Competing mandates
• Proposed solutions
• PDA or computer entry prompts

43
Act On

• Multiple other standards of care or quality
  measures are more important priorities
• Yes 47/76 (61.8)
• No 29/76 (38.2)
• Steroid administration is not a QI measure and is
  therefore less important than other clinical
  performance measures
• Yes 21/75 (28.0)
• No 54/75 (72.0)
• ED overcrowding precludes the timely recognition
  of ABM necessary to safely administer steroids
• Yes 23/76 (30.3)
• No 53/76 (69.7)

44
Act On

• ED overcrowding reduces my confidence in my
  ability to recognize ABM and permit the safe
  administration of steroids
• Yes 17/76 (22.4)
• No 59/76 (77.6)
• Remembering to order steroids in suspected ABM
  while caring for multiple other sick patients in
  the ED is impossible
• Yes 4/76 (5.3)
• No 72/76 (94.7)
• No order set exists to facilitate steroid
  administration to suspected ABM patients
• Yes 61/72 (84.7)
• No 11/72 (15.3)

45
Agree

• Leaks in this leg of the Pipeline include
• Information overload
• Competing influences
• Proposed solutions
• Community EBM education
• Anticipate couterarguments

46
Agree

• Lacking of an established statement listing
  steroids in ABM as standard of care (e.g. ACEP
  guideline), I fear increased malpractice risk if
  I use steroids for this indication
• Yes 12/76 (15.8)
• No 64/76 (84.2)

47
Summary and Subgroup Analysis

• Though leaks in the pipeline were identified at
  each step, subgroup analysis of the two groups
  (steroid users vs. non-users) highlighted two
  areas in particular
• Awareness
• Steroid users were more aware of supporting
  research, appropriate doses and timing of
  steroids, and reported having been taught that
  steroids are beneficial in ABM when compared with
  non-users
• Acceptance
• Steroid users reported less doubt within their
  physician and nursing peer groups when compared
  with non-users

48
External Validation

• In order to assess whether or not this Knowledge
  Translation Gap exists in other institutions, a
  similar survey was sent out to a number of EM
  programs across the country, including both
  academic and community centers
• 35 of 65 (53.8) responded and data analysis was
  performed
• 18 of 35 (51.4) almost always give steroids to
  adult patients with suspected ABM
• Reasons for integrating steroids into the
  management pathway in ABM?
• The supporting literature is indisputable (73.7)
• My clinical instructors taught me that steroids
  are standard of care (42.1)
• 17 of 35 (48.6) do not almost always give
  steroids

49
What Explanations Were Given as to Why Steroids
are not Routinely Used in ABM?

• Awareness
• I am unaware of the correct dose of steroids
  35.7
• Unaware of evidence supporting use of steroids in
  ABM 14.3
• Acceptance
• Research evidence is inconclusive 42.9
• Able
• Diagnostic delays impede the timely
  administration of steroids in ABM 14.3
• Act On
• I simply forget to administer steroids in the
  rare ABM patient 35.7
• Steroids are not part of my hospitals meningitis
  treatment algorithm or standardized orders
  21.4

50
What Solutions Were Proposed by Outside
Institutions to Fix These Leaks in the Pipeline?

• Awareness
• Journal Club 82.4
• Didactic programs within the institution 35.3
• Act On
• Order sets within the ED 5.9
• Adhere
• Quality Improvement teams review performance
  17.6
• Individual responders noted that their awareness
  came from
• Attendings distributing articles
• LLSA (ABEM article)
• BEEM

51
What Barriers Were Encountered While
Incorporating the Use of Steroids in ABM?

• In the hectic ED environment, I simply forget to
  administer steroids in ABM (55.6)
• Difficulty distinguishing between viral and
  bacterial meningitis before LP (50.0)
• Delays in diagnosis CNS infection (38.9)
• Concerns about steroid-related adverse effects
  (27.8)
• Uncertainty about the timing of steroids relative
  to antibiotics and CSF results (27.8)
• Difficulty distinguishing viral from bacterial
  meningitis after the LP (16.7)
• Consult services disagreed with steroids in ABM
  (16.7)
• Skepticism from my EM colleagues (11.1)
• Inability to understand the evidence supporting
  this therapy (5.6)

52
Conclusions

• Wash U EM does not differ significantly from our
  national sampling for
• Awareness of supporting literature (14 vs 18
  unaware)
• Timing of steroids relative to antibiotics (28
  in both groups unaware)

53
Conclusions, continued

• BJH physicians were more likely to accept the
  supporting literature (29 BJH vs. 43 outside
  institutions felt data inadequate), but less
  likely to be using steroids in meningitis (41
  BJH vs. 51 outside institutions currently using
  steroids in suspected adult meningitis)
• Dosing at BJH often inappropriate
• 26 with antibiotics
• 9 after antibiotics

54
Conclusions

• Potential faucets for the largest leaks include
• Tonights literature review (Awareness/Acceptance)
  
• Steroids into meningitis order set (Able/Act On)
• Frequent reminders of steroids in meningitis via
• Didactics
• LLSA
• QI reviews
• Incorporation into the standard of care
• Textbooks
• Narrative reviews
• Medical student/GME educational models

55
The Biggest Hurdle

• Atypical disease presentation and delayed
  diagnoses
• Delayed antimicrobials increase bacterial
  meningitis related morbidity and mortality
• Viral and bacterial CNS infections present with
  similar signs/symptoms
• CSF findings available in the ED cannot reliably
  distinguish viral and bacterial meningitis
• Steroids may be harmful to non-bacterial CNS
  infections such as TB, HIV, and HSV encephalitis

56
My Recommendation

• Steroids should be administered 15-minutes prior
  to antibiotics if
• Positive CSF gram stain
• Clear petechiae rash in conjunction with fever,
  headache, or altered mental status
• Purulent CSF
• CSF WBC gt 1000 cells/mm3
• High clinical suspicion of bacterial meningitis

57
A Knowledge Translation Journal Club such as
weve discussed tonight is superior to our
typical JC.

1. Yes
2. No
3. Uncertain

58
I would like to have another Knowledge
Translation project next year.

1. Yes
2. No

59
Tonights discussion changed my opinion about the
use of steroids in meningitis.

1. Yes, from use to dont use
2. Yes, from dont use to use
3. No

60
Washington Universitys EM program should
purchase the Turning Point system for routine use
in our lectures and Journal Club

1. Yes
2. No

61
Future Directions

• Following this Journal Club, we will analyze
  management of potential meningitis cases
  presenting to the ED
• Annual Knowledge Translation projects would
  continue to help shorten the lag time needed to
  bring high quality evidence into clinical
  practice by identifying pertinent leaks in the KT
  Pipeline