Risk Classification in Community-acquired Pneumonia

1
Risk Classification in Community-acquired
Pneumonia

• Thomas M File, Jr MD MACP FIDSA FCCP
• Chair, Infectious Disease Division
• Summa Health System
• Akron, Ohio
• Professor of Internal Medicine,
• Chair Infectious Disease Section
• Northeast Ohio Medical University
• Rootstown, Ohio

2
(No Transcript)
3
Community-acquired Pneumonia (CAP)

• Leading cause of morbidity and mortality
• No. I cause due to infection
• 5-6 million cases/year
• gt 75 treated as outpatients
• Approx. 1 million admissions/year
• 40 one year mortality Kaplan et al. Arch Intern
  Med 2003 163 317-323)
• 50 mortality at 30 months (Bordon et al. chest)
• Cost of treating CAP exceeds 17 billion/year

File T. Lancet 2003 File and Tan JAMA 2005 File
T and Marrie T Postgrad Med. 2010
4
Patient Stratification for Management of
Community-acquired Pneumonia

• Patient stratification valuable for optimal
  management
• Helpful in
• Assessing NEED for antimicrobial therapy (i.e.,
  viral vs bacterial etiology)
• Determining Severity of illness (site of care
  for CAP)
• Prognosis/Outcomes
• Predicting likely pathogen/Risk of Resistance
• Guidelines recommend stratifying patients
  according to risk factors1,2

1. Mandell LA et al. Clin Infect Dis.
200744(suppl 2)Lim et al.. BTS, Thorax 2009
64 Suppl 3.
5
CAP Case

• 66 Y/O MALE
• Smoke, Diabetes, CHF
• Treated with macrolide for sinusitis 8 weeks
  ago
• Headache, Fever, Cough for 3 days, New Confusion
• T-38.60 C P-110 RR-28 Ausc-rhonchi RLL
• O2 sat-92 Room Air
• Should he be admitted?

6
Patient Stratification CAP Site of Care

• Variability in hospital admission rates
• Clinicians use inconsistent criteria
• Likely overestimate need for admission
• Several prediction tools of severity of illness
  of CAP
• Legitimate decision aids
• Two best studied PSI and CURB 65
• Assessment of mortality risk
• PSI best studied and only prognostic rule shown
  to safely reduce proportion of low risk patients
  hospitalized for care
• PSI may oversimplify some predictor variables (no
  degree of abnormality) and based on 19 variables
• CURB 65 simpler better for mortality does not
  include most comorbidities

• Marrie T. UptoDate 2008 Aujesky and Fine Clin
  Infect dis. 2008 47 S133
• Atlas et al. Arch Intern Med. 1998 Marrie et
  al. JAMA 2000 Yealy et al Ann Intern Med. 2005

7
Risk Stratification Site of Care Decision

• Determines
• Cost of care
• Intensity of diagnostic testing
• Empiric choice of antibiotics
• Advantages of outpatient therapy
• Cost
• Patient preference
• Faster convalescence and avoidance of nosocomial
  complications
• Science and Art
• Mortality prediction rules (PSI, CURB-65)
• Social circumstances
• Co-existing conditions

8
Pneumonia PORT Prediction Rule for Mortality
Risk Assessment (PSI)
STEP 1
STEP 2
Yes
Is the patient gt50 years of age?
Class II(?70 points)
Assign points for Demographic variables Comorbid
conditions Physical observations Laboratory and
radiographic findings
No
Does the patient have any of the following
coexisting conditions Neoplastic disease
congestive heart failure cerebrovascular
disease renal disease liver disease
Class III(7190 points)
Yes
Class IV(91130 points)
No
Yes
Does the patient have any of thefollowing
abnormalities Altered mental status pulse
?125/min respiratory rate ?30/min systolic
blood pressure lt90 mm Hg temperature lt35ºC or
?40ºC
Class V(gt130 points)
No
Class I
Fine MJ, et al. N Engl J Med. 1997336243-50.
PSIPneumonia Severity Index
9
Prediction Rule Step 2 Algorithm

• Pt Characteristic
  Points
• Age No. of years (-10 for female)
• Cancer 30
• Liver disease 20
• CHF, CVD, Renal disease 10
• RR gt30/min, SBP lt90 mmHg, Confusion 20
• Temp lt35ºC, gt50ºC 15
• Pulse, beats/min 10
• BUN Sodium lt130 mmol/l 20
• Glucose gt250 mg/dl Hct lt 30 10
• pO2 lt 60 mmHg 10

10
Prediction Rule Risk Categories
Total Points Class Mortality How to Treat
I 0.1 Outpatient
? 70 II 0.6 Outpatient
71-90 III 0.9-2.8 Brief hospital observation
91-130 IV 8.2-9.3 Inpatient
gt130 V 27.0-29.2 Inpatient ICU
Risk categories according to two validation
cohorts (38,039 inpatients and 2287 in- and
outpatients) Mortality 1 year after hospital
discharge 33 for patients gt65 years old
(Kaplan V, et al. Arch Intern Med 2003
163317-23.)
Fine MJ, et al. N Engl J Med.
1997336243-50.
11
PSI Amended Algorithm

• 1. Assessment of preexisting conditions that
  compromise homecare hypoxemia severe social
  or psychiatric problems inability to take oral
  meds
• 2. Calculation of PORT Severity Index
• 3. Clinical judgment regarding overall health of
  the patient and suitability for home care.
  Clinical judgment should supersede the severity
  of illness

Metlay and Fine Ann Intern Med. 2003 138
109-118
12
Applying the CURB-65 Rule
Group 1 Mortality Low (1.5) (n324, died5)
Treatment Options
CURB-65 Score
Likely suitable for home treatment
0 or 1
Any of Confusion Urea gt7 mmol/l Respiratory
Rate 30/min Blood pressure (SBP lt90 mmHg or DBP
60 mm Hg) Age 65 years
Consider hospital supervised treatment Options
may include Short stay inpatient Hospital-super
vised outpatient
Group 2 Mortality Intermediate (9.2) (n184,
died17)
2
Manage in hospital as severe pneumonia Assess for
ICU admission especially if CURB-65 score 4 or 5
3
Group 3 Mortality High (22) (n210, died47)
Lim WS, et al. Thorax. 200358377-82.
13
Practical Severity Assessment Model(without
blood test)
Lim et al. Thorax 2003 58 377-82
14
PSI, CRB-65 versus CURB-65
CRB-65 Score 30-day Mortality ()
0 0
1 4.1
2 18.7
3 43.5
4 54.6
PSI (area under the curve (AUC) 0.888
CURB-65 AUC 0.870 --------- CRB-65 AUC 0.864

• CRB-65 omits blood urea measurement
• Applicable to office-based settings
• Scores of 0 home treatment, 1
  hospital-supervised treatment, 2
  hospitalization

Capelastegui A, et al. Eur Respir J.
200627151-57.
15
Predicting outcomes using CRB-65
From CAPO database 2926 patients
CRB-65 0 1 2 3 4 P value
TCS (days) 4.0 4.7 5.2 6.4 6.9 lt0.0001
LOS (days) 7.0 8.5 10.0 12.4 13.6 lt0.0001
Mortality 6.1 9.0 15 50 88.9 lt0.0001
CAPO Community acquired pneumonia organization
TCSTime to clinical stability LOSLength of stay
Arnold F. et al. Community Acquired Pneumonia
Organization 2006
16
CURB-65-as continuous variable

• Traditional CURB assessment is binary
• 62 oriented male, BUN 44, BP 95/65, RR 28
  CURB-65 of 1 (mortality of 2)
• Continuous Variable
• Same patient with predicted mortality of 14

Jones BE et al. Chest 2011 140 156-63
17
Criteria for ICU Admission

• Major Criteria
• Invasive mechanical ventilation
• Septic shock with the need for vasopressors
• Minor Criteria
• Confusion/disorientation
• Blood urea nitrogen 20 mg/dL
• Respiratory rate 30 breaths/min
• Hypotension requiring aggressive fluid
  resuscitation
• PaO2/FiO2 ratio 250
• Multilobar infiltrates
• WBC lt4000 cells/mm
• Platelet count lt100,000 cells/mm
• Core temperature lt36oC

Direct admission to ICU recommended if 1 major
(strong recommendation) or 3 minor criteria
(moderate recommendation)
Mandell L, et al. Clin Infect Dis. 200744 (Suppl
2)S27-72.
18
Prediction for Severe CAP (SCAP)

• Major Criteria Points
• pH lt 7.30 13
• Systolic pressure lt 90 mm Hg 11
• Minor Criteria
• Resp rate gt 30 9
• Blood urea nitrogen gt 30 mg/dL 5
• Respiratory rate 30 breaths/min
• Altered mental status 5
• PaO2/FiO2 ratio 250 6
• Age gt 80 5
• Multilobar/bilateral infiltrates 5

SCAP 1 Major or 2 minor Mortality based on
points 1-9 2.4 10-19 9.26 20-29 42.37 30
75
Espana PD et al. Am J Resp Crit Care Med. 2006
174 1249-56
19
Pitt Bacteremia Score (PBS)

• Criteria Points
• Fever
• 35C or 40C 2
• 35.1C-36C or 39.0-39.9 1
• 36.1-38.9C 0
• Hypotension 2
• Mech Vent 2
• Cardiac Arrest 4
• Mental status
• Alert 0
• Disoriented 1
• Stuporous 2
• Comatose 4

SEVERE gt 4
Paterson D et al. Ann Intern Med 2004 140 26-32
20
SMART-COP predicting need for ICU
Score 3-4 low risk 5-6 high risk (33) 7
Very High Risk (66)
PG Charles et al. Clin Infect Dis. 2008 47
375-84
21
Modified ATS Score

• Major Criteria
• Mech Vent
• Septic shock
• Minor Criteria
• Systolic BP lt 90 mmHg
• Multilobar (gt lobes) involvement
• PaO2/FiO2 ratio lt 250

Severe 1 Major or 2 minor
Angus DC et al. Am J Respir Crit Care Med 2002
166 717-723
22
Late admission to the ICU is associated with
higher mortality
Further studies should examine variables that
may allow clinicians to determine patients who
will have a late clinical failure, and processes
of care that may reduce the need for these
late transfers. An effective tool for evaluation
of disease progression will identify patients who
are at risk for clinical deterioration.
EICUAearly ICU admission LICAUAlate ICU
admission
Restrepo MI et al. Chest 2010 137 552-557
23
Validation of IDSA/ATS minor criteria for ICU
admission
The IDSA/ATS 2007 minor criteria are not perfect
and require additional impact and
validation analyses but seem to be the scoring
system closest to achieving these goals so far
we conclude that they accurately predict
requirement for MV/VS, ICU admission, and
30-day mortality in patients with CAP.
Chalmers JD et al. Clin
Infect Dis. 2011 53503-11
24
Severity scoring systems in bacteremic
pneumococcal pneumonia implications for ICU care
Feldman C et al. Clin Microbiol Infect 2009 15
850-857
25
CAP Case

• 66 Y/O MALE
• Smoke, Diabetes, CHF
• Treated with macrolide for sinusitis 8 weeks
  ago
• Headache, Fever, Cough for 3 days, New Confusion
• T-38.60 C P-110 RR-28 Ausc-rhonchi RLL
• O2 sat-92 Room Air
• Patient is admitted (PSI IV, CURB-65 2 mortality
  8-9)
• What antimicrobial therapy?

26
Empiric Therapy in CAP IDSA/ATS
26
Healthy Outpatient Outpatient at Risk for DRSP Inpatient, non-ICU Inpatient, ICU
Macrolide OR Doxycycline Respiratory fluoroquinolone OR Beta-lactam plus macrolide Respiratory fluoroquinolone OR Beta-lactam plusmacrolide Beta-lactam plus azithromycin OR Beta-lactam plus fluoroquinolone (Special concerns Pseudomonas, CA-MRSA)
Includes healthy patients in regions with high
rates of macrolide resistance. levofloxacin
750 mg, moxifloxacin 400 mg, gemifloxacin 320
mg. ceftriaxone, cefotaxime, amp/sulbactam,
ertapenem Treatment of Pseudomonas
(anti-pseudomonal beta-lactam regimen) or MRSA
(vancomycin or linezolid) is the main reason to
modify standard therapy for ICU patients.
IDSA Infectious Diseases Society of America
ATSAmerican Thoracic Society ICU intensive
care unit
Mandell L, et al. Clin Infect Dis.
200744(Suppl 2)S27-S72.
27
BTS Guidelines for CAP-2009 UpdateThorax 2009
64 Suppl III
Severity (CURB65) Site Preferred Alternative
Low (0-1) Home Amoxicillin 500 mg tds orally Doxycyline or Clarithromycin
Low (0-1) Hospital for other than severity Amoxicillin 500 tds orally Doxycyline or Clarithromycin
Moderate (2) Hospital Amox .5-1G Clarithromycin orally OR (BenzylPCN or Amox) Clarithro IV Doxycyline or Levofloxacin (500 QD) or Moxifloxacin (400 QD)
High (3-5) Hospital (consider ICU) Co-amox/clav clarithromycin (if legionella suspected -Levo) BenzylPCN (Levo or ciprofloxacin) OR Cefuroxime or ceftriaxone Clarithro (if legionella suspected -Levo)
28
Use of Procalcitonin for Stratification of
Antimicrobial Use for RTIs
PCT lt 0.1 ug/ml Bacterial Infection VERY UNLIKELY NO ANTIMICROBIALS Consider repeat 6-24hrs based on clinical status
PCT 0.1-0.25 ug/ml Bacterial infection UNLIKELY NO ANTIMICROBIALS Use of ABX based on clinical status (unstable) judgment
PCT gt 0.25-0.5 ug/ml Bacterial infection LIKELY YES ANTIMICROBIALS Repeat PCT day 3, 5, 7 (for Duration)
PCT gt 0.5 ug/ml Bacterial infection VERY LIKELY YES ANTIMICROBIALS CONSIDER STOP ABX when 8090 decrease if PCT remains high consdier treatment failure
File TM Jr. Clin Chest Med. 2011 modified from
Schuetz P. et al. Eur Respir J 201137(2)
38492.
29
Risk Classification in Community-acquired
Pneumonia Summary

• Multiple severiy assessment tools based on
  different outcomes have been developed to risk
  stratify patients with CAP
• Clinicians should combine clinical judgment with
  risk assessment tools to consider management
  decisions to provide optimal outcomes for our
  patients