Pneumonia Management & Chris Fee, MD Assistant Clinical Professor UCSF Department of Emergency Medicine The Old Man s Friend (And the Ire of Many an ED QI Director) 7th leading cause of death in US 915,000 cases/year in age >65 in US Mortality rate stable $8.4-10 billion/year Objectives Diagnosis Disposition Decision How to Meet the Core Measures While Doing the Right Thing for Your Patients Antibiotic Timing Antibiotic Selection Case Identification Fighting the Good Fight Terminology Community-Acquired Pneumonia (CAP)* Healthcare-Associated Pneumonia (HCAP) Hospital-Acquired Pneumonia (HAP) Ventilator-Associated Pneumonia (VAP) *IDSA/ATS Guidelines 2007 ATS/IDSA Guidelines 2005 1
IDSA/ATS CAP Guidelines Do Not Apply To Immunocompromised Transplant recipients Chemotherapy or long-term (>30 days) high-dose corticosteroids Congenital/acquired immunodeficiency HIV with CD4 < 350 cells/mm 3 Children 18 years old IDSA/ATS Grading of Evidence Strength of recommendation Strong Moderate Weak Level of evidence I (high): well-conducted RCTs II (moderate): well-conducted, non RCTs III (low): case studies, expert opinion Diagnosis of CAP Suggestive Clinical Features* (cough, fever, sputum, pleuritic chest pain) + Demonstrable Infiltrate by CXR (CT more sensitive,? significance with negative CXR) Physical exam: less sensitive/specific than CXR *Clinical features and exam may be lacking in altered/elderly patients Moderate/III 2
Slide on CXR vs CT Include CXR & CT images from JAMA article Scarlet letter Diagnosis CXR vs CT Disposition Decision Severity of illness scores -OR- Prognostic models identify patients for outpatient treatment* Strong/I *As always physician judgment supercedes scores Strong/II Confusion Uremia (>20mg/dL) Disposition Decision CURB-65 Respiratory 30 Blood pressure (systolic<90, diastolic 60) 65 years old # Factors 0 1 2 3 4 5 30d Mortality 0.7% 2.1% 9.2% 14.5% 40% 57% Site of Care Outpatient Inpatient Wards ICU Moderate/III Confusion Uremia (>20mg/dL) Disposition Decision CURB-65 Respiratory 30 Blood pressure (systolic<90, diastolic 60) 65 years old # Factors 0 1 2 3 4 5 30d Mortality 0.7% 2.1% 9.2% 14.5% 40% 57% Site of Care Outpatient Inpatient Wards ICU Moderate/III 3
Disposition Decision Pneumonia Severity Index Points <51 51-70 71-90 91-130 >130 PSI Class I II III IV V 30d Mortality (%) 0.1-0.4 0.6-0.7 0.9-2.8 8.2-9.3 27.0-31.1 Re-hosp rate (%) 5.1 8.2 16.7 20.0 0 (n=1) Site of Care Outpatient Obs unit/short stay Inpatient ward ICU Disposition Decision Pneumonia Severity Index Points <51 51-70 71-90 91-130 >130 PSI Class I II III IV V 30d Mortality (%) 0.1-0.4 0.6-0.7 0.9-2.8 8.2-9.3 27.0-31.1 Re-hosp rate (%) 5.1 8.2 16.7 20.0 0 (n=1) Site of Care Outpatient Obs unit/short stay Inpatient ward ICU Age (in years) Age Female sex -10 Nursing home resident 10 Non-skin cancer (active/dx 1 year) 30 Liver disease (cirrhosis/chronic dx) 20 CHF 10 Cerebrovascular disease 10 Renal disease 10 AMS 20 RR 30/minute 20 SBP 90 mmhg 20 Temp 35ºC or 40ºC 15 Pulse 125/minute 10 Arterial ph < 7.35 30 BUN 30 mg/dl 20 Na < 130 meq/l 20 Glucose > 250 mg/dl 10 Hematocrit < 30% 10 Arterial po2 < 60 mmhg 10 Pleural effusion 10 http://pda.ahrq.gov/clinic/psi/psicalc.asp Disposition Decision ICU Admission-Severe CAP Septic shock or respiratory failure 3 minor criteria for severe CAP RR 30 breaths/min P a O 2 /F i O 2 250 Multilobar infiltrates Confusion/disorientation BUN 20 mg/dl Leukopenia Thrombocytopenia Strong/II Hypothermia (<36ºC) Hypotension Moderate/II Disposition Decision SMART-COP ICU Admission? Systolic BP < 90 mmhg 2 Multilobar CXR involvement 1 Albumin < 3.5 g/dl 1 RR (age adjusted) <50 yrs: 25 resps/min 50 yrs: 30 resps/min Tachycardia 125 bpm 1 Confusion (new onset) 1 Oxygen low (age adjusted) <50yrs: PaO2<70, SaO2<93%, PaO2/FiO2<333 50 yrs: PaO2<60, SaO2<90%, PaO2/FiO2<250 ph < 7.35 2 1 1 2 2 Interpretation 0-2 low risk 3-4 mod (1/8) 5-6 high (1/3) 7 very high (2/3) of need for intensive respiratory or vasopressor support Charles PG, et al. CID 2008 4
PSI Class 2 (3) CURB-65 < 2 SMART-COP 2 Disposition Decision The Bottom Line Yes Hypoxia (SaO2 <90%) Active Coexisting Condition Unable to Tolerate POs Failed Outpatient Therapy Insecure Social Setting No PSI Class = 3 CURB-65 = 2 SMART-COP = 3-4 PSI Class = 4 CURB-65 2-3 SMART-COP = 5-6 PSI Class = 5 CURB-65 =3-5 SMART-COP 7 Antibiotic Timing Antibiotic Selection Case Identification Home Brief Stay SNF Home IV Home Nursing Close follow-up Inpatient Wards ICU Joint Commission/CMS Core Measures Compliance Oxygenation Assessment (PN-1) Blood cultures within 24 hours of arrival for ICU admits (PN-3a) ED blood cultures prior to 1st antibiotic (PN-3b) Time to first antibiotic dose (TFAD) (PN- 5b/c) Initial antibiotic selection (PN-6a/b) Do you obtain blood cultures for routine hospitalized CAP? A. Yes B. No C. Sometimes 5
Do you obtain blood cultures for routine hospitalized CAP? A. Yes B. No C. Sometimes D. I do now (thanks to the core measures) Most Common CAP Etiologies Outpatient Inpatient (non-icu) Inpatient (ICU) S. pneumoniae S. pneumoniae S. pneumoniae M. pneumoniae M. pneumoniae S. aureus H. influenzae C. pneumoniae Legionella spp. C. pneumoniae H. influenzae Gram-negative bacilli Respiratory viruses Legionella spp. H. influenzae Aspiration Respiratory viruses Most Common CAP Etiologies Outpatient Inpatient (non-icu) Inpatient (ICU) S. pneumoniae S. pneumoniae S. pneumoniae M. pneumoniae M. pneumoniae S. aureus H. influenzae C. pneumoniae Legionella spp. C. pneumoniae H. influenzae Gram-negative bacilli Respiratory viruses* Legionella spp. H. influenzae Aspiration Respiratory viruses* *Influenza A/B, adenovirus, RSV, parainfluenza Etiologies Not Empirically Covered Outpatient Inpatient (non-icu) Inpatient (ICU) Respiratory viruses* Legionella spp. S. aureus Aspiration Respiratory viruses* Legionella spp. Gram-negative bacilli *Influenza A/B, adenovirus, RSV, parainfluenza 6
Recommended* ONLY IF results would alter standard management AND suspected by clinical/epidemiologic clues *Optional for outpatients with CAP Strong/II Moderate/III Condition (Epidemiology) Alcoholism COPD +/- smoking Aspiration Lung abscess Exposure to bat or bird droppings Exposure to birds Exposure to rabbits Exposure to farm animals or parturient cats HIV infection (early) HIV infection (late) Hotel/cruise ship within 2 weeks Travel to/residence in SW US Travel to/residence in SE/E Asia Influenza active in community Cough > 2 wks with whoop/posttussive vomiting Structural lung disease (bronchiectasis) Injection drug use Endobronchial obstruction Bioterrorism Risk Factors for Specific Pathogens Pathogen S. Pneumoniae, oral anaerobes, K. pneumoniae, Acinetobacter spp. TB H. influenzae, P. aeruginosa, Legionella spp, S. pneumoniae, M. catarrhalis, Chlamydophila pneumoniae Gram-negative enteric pathogens, oral anaerobes CA-MRSA, oral anaerobes, endemic fungal pneumonia, TB, atypical mycobacteria Histoplasma capsulatum Chlamydophila psittaci (if poultry: avian influenza) Francisella tularensis Coxiella burnetti (Q fever) S. pneumoniae, H.influenzae, TB Above + Pneumocystis jirovecii, Cryptococcus, Histoplasma, Aspergillus, atypical mycobacteria, P. aeruginosa, H. influenzae Legionella spp Coccidioides spp, Hantavirus Burkholderia pseudomallei, avian influenza, SARS Influenza, S. pneumoniae, S. aureus, H. influenzae Bordetella pertussis P. aeruginosa, Burhkholderia cepacia, S. aureus S. aureus, anaerobes, TB, S. pneumoniae Anaerobes, S. pneumoniae, H. influenzae, S. aureus B. anthracis (anthrax), Yersinia pestis (plague), Francisella tularensis (tularemia) Nasal Wash for Viral Pathogens If concern for influenza, RSV, adenovirus, parainfluenza Blood Cultures True positives 5-14% Most commonly S. pneumoniae Negligible impact on management False positives Prolonged hospital stays More vancomycin use 7
ICU admissions Cavitary lesions Leukopenia Alcohol abuse Blood Cultures Recommended For Chronic severe liver disease Asplenia (anatomic/functional) Pleural effusion Do you obtain sputum for Gram stain & culture for routine hospitalized CAP? A. Yes B. No C. Sometimes IDSA/ATS 2007 Sputum Gram Stain & Cultures Yield = Highly Variable Only 14% of 1669 hospitalized CAP pts had adequate specimen with predominant morphotype on Gram stain Specimen collection Transport Rapid processing Cytologic criteria Absence of prior antibiotic therapy Skill in interpretation Sputum Gram Stain & Cultures Recommended For ICU admission Failure of outpatient antibiotic therapy Cavitary lesion Active alcohol abuse Severe obstructive/structural lung disease Pleural effusion Moderate/I IDSA/ATS 2007 8
If you had access to urinary antigen testing for Legionella and/or pneumococcus, would you use it? A. Yes B. No Urinary Antigen Testing Pneumococcal UAT Rapid (15 mins) Sens 50-80%, Spec >90% Remains + after antibiotic therapy initiated $30 per specimen Moderate/II Urinary Antigen Testing Legionella serotype 1 UAT Causative serotype in 80-95% Sens 70-90%, Spec 99% + on day 1, remains + for weeks Indication (Clinical) Blood Culture Sputum Culture Legionella UAT Pneumococcal UAT Other ICU Admission X X X X Endotracheal aspirate,? bronchoscopy, nonbronchoscopic BAL Failure of outpatient therapy X X X Cavitary lesions X X Fungal and TB cultures Leukopenia X X Alcohol abuse X X X X Chronic severe liver disease X X Severe obstructive/structural lung disease Targeted Testing Asplenia (anatomic of functional) X X X Recent travel (within 2 weeks) X See prior Table Pleural effusion X X X X Thoracentesis and pleural fluid cultures Moderate/II 9
The Bottom Line The Culture Wars Indication (Clinical) ICU Admission Cavitary lesions Leukopenia Alcohol abuse Chronic severe liver disease Asplenia (anatomic of functional) Pleural effusion Blood Culture X X X X X X X TJC/CMS measure % ICU admits with blood cultures within 24 hrs of arrival Severity of illness/prognostic scoring to predict The UCSF experience 1. ED MD feels blood cultures not indicated 2. Antibiotics administered 3. Admit team orders blood cultures (!@%#*?) 4. ED labeled as giving antibiotics prior to obtaining blood cultures Time to First Antibiotic Dose 1997 TFAD 8 hrs: lower 30d mortality 2004 TFAD 4 hrs: reduced in-hospital mortality (6.8 vs 7.2%) reduced 30d mortality (11.6 vs 12.7%) decreased length of stay Time to First Antibiotic Dose Controversies Higher mortality for TFAD <1 hr in both studies 2 smaller studies found no association between TFAD and outcomes No improvement if on outpatient antibiotics 10
Time to First Antibiotic Dose 1997: Medicare National Pneumonia Project endorsed TFAD 8 hrs as CAP quality measure 2002: MNPP endorsed TFAD 4 hrs 2002: TJC/CMS choose TFAD 4 hrs as core measure & publicly report www.hospitalcompare.hhs.gov 2003: IDSA endorsed 4 hr TFAD Time to First Antibiotic Dose The Bottom Line Time to First Antibiotic Dose Effective April 2008: TFAD changed to 6 hrs New data element: diagnostic uncertainty May exclude pts from the TFAD measure Antibiotic Therapy 2007 IDSA/ATS CAP guidelines: For patients admitted through the ED, the first antibiotic dose should be administered while still in the ED Moderate/III 11
Moxifloxacin Gemifloxacin Levoflox (750 mg) Antibiotic Therapy Outpatient Azithromycin Clarithromycin Erythromycin Antibiotic Therapy Inpatient (non-icu) Previously healthy and no risk factors for DRSP* infection Presence of comorbidities or risks for DRSP* High-dose amox (1gm TID) Amox-clavulanate (2gm BID) Ceftriaxone, cefpodoxime, cefuroxime (500mg BID) Region with >25% macrolideresistant S. pneumoniae A. Macrolide (Strong/I) B. Doxycycline (Weak/III) A. Respiratory quinolone (Strong/I) B. β-lactam* + macrolide (Strong/I) Doxycyline intead of macrolide (II) Consider alternate agents in previously healthy pts DRSP = drug resistant S. pneumoniae Risks for DRSP = age <2 or >65 yrs, β-lactam therapy < 3 months*, alcoholism, medical comorbidities, immunosuppressive illness or therapy, exposure to child in daycare Respiratory quinolone (Strong/I) β-lactam* + macrolide (Strong/I) Doxycyline as alternative to macrolide (III) Respiratory quinolone alone if PCN allergic *cefotaxime, ceftriaxone, ampicillin, ertapenem The Bottom Line Ceftriaxone + Doxycycline Levofloxacin if PCN allergic Antibiotic Therapy Inpatient (ICU) β-lactam* + azithromycin (II) or fluoroquinolone (I) (Strong) *cefotaxime, ceftriaxone, ampicillin-sulbactam PCN allergic: fluoroquinolone + aztreonam The Bottom Line Ceftriaxone + Azithromycin (IV) Levofloxacin if PCN allergy Pseudomonas CA-MRSA Antibiotic Therapy Inpatient (ICU) Special Considerations Antipneumococcal, antipseudomonal β-lactam* + A. cipro or levofloxacin (750 mg) or B. aminoglycoside & azithromycin or C. aminoglycoside & antipneumococcal fluorquinolone (Moderate/III) Add vancomycin or linezolid (Moderate/III) Pseudomonas risk factors: COPD, late HIV, structural lung dz The Bottom Line Pseudomonal risks are only indication for Piperacillin/Tazobactam!!! 12
*HCAP Includes Hospitalized 2 days within 90 days SNF or long-term care facility IV antibiotic therapy, chemotherapy, or wound care within 30 days Hospital or hemodialysis clinic within 30 days Increased risk for multi-drug resistant pathogens Outpatient CAP Inpatient CAP (non-icu) HCAP Etiology Inpatient CAP (ICU) HCAP S. pneumoniae S. pneumoniae S. pneumoniae S. pneumoniae M. pneumoniae M. pneumoniae S. aureus H. influenzae H. influenzae C. pneumoniae Legionella spp. MSSA C. pneumoniae H. influenzae Gram-negative bacilli Gram-negative bacilli Respiratory viruses Legionella spp. H. influenzae P. aeruginosa Aspiration Respiratory viruses K. pneumoniae Acinetobacter spp. MRSA Legionella spp. HCAP Etiologies Not Empirically Covered Outpatient CAP Respiratory viruses* Inpatient CAP (non-icu) Inpatient CAP (ICU) HCAP Legionella spp. S. aureus MSSA Aspiration Legionella spp. Gram-negative bacilli Respiratory viruses* Gram-negative bacilli P. aeruginosa K. pneumoniae Acinetobacter spp. MRSA Legionella spp. Case Identification 1. Random selection of hospitalized patients discharged with pneumonia (ICD-9) 2. Review ED physician chart *Influenza A/B, adenovirus, RSV, parainfluenza 13
Case Identification 1. Random selection of hospitalized patients discharged with pneumonia (ICD-9) Did they really have pneumonia? Did not require any objective supporting data Case Identification 2. Review ED physician chart - Initially included in measure if any mention of pneumonia on the ED chart (medical decision making, differential, or final ED dx) Case Identification Review ED physician chart Exclude if No ED pneumonia diagnosis Beware: infiltrate, consolidation, possible, rule out On outpatient antibiotics Earliest documented time = time zero Common Pitfalls CAP as final ED diagnosis, no antibiotic CT to r/o PE finds CAP, antibiotics delayed Possible, infiltrate, consolidation, rule out in final ED diagnosis Delay from antibiotic order to administration Piperacillin/tazobactam!!! 14
Fight The Power (A User s Guide) Lack of Representation Grumbling Does No Good Do Something About It Research (Pro-active vs Reactive) Advocacy Collaboration Advocate for Your Patients and EM Like It Or Not Measures proposed/solicited Expert panels in NQF or AMA-PCPI Performance measure clearing houses Board approval CMS/Joint Commission select from this list 15
Take Home Points Utilize illness severity/prognostic scores to aid disposition decisions Targeted etiologic testing only Appropriate antibiotic selection trumps early antibiotics Beware HCAP--more to come Beware the common pitfalls related to the core measures Do the right thing for your patients!!! Questions? 16