Bradley A. Sharpe, M.D. Associate Professor Medicine Department of Medicine UCSF sharpeb@medicine.ucsf.edu a. An ailment that often leads to suffocation and death. b. A friend of the aged. c. A common and mortal disease which can be diagnosed by simple observation and percussion of the chest. d. Bad. Really bad. "Pneumonia may well be called the friend of the aged. Taken off by it in an acute, short, not often painful illness, the old man escapes those cold gradations of decay so distressing of himself and to his friends. -- William Osler, M.D., 1898 1
Brad, pneumonia sucks. -- Mary R. Sharpe November 2008 Background Etiology Diagnosis Treatment Prevention Guidelines for Community-Acquired Pneumonia IDSA/ATS Consensus Guidelines 2007 (IDSA = Infectious Disease Society of America) (ATS = American Thoracic Society) BTS: British Thoracic Society Updated Literature Review 2
Will not talk about healthcare-associated pneumonia (HCAP) Will not discuss admission decision (complex) Syllabus vs. all slides Background Etiology Diagnosis Treatment Prevention Community-Acquired Pneumonia 5 million cases/year in the U.S. 80% of CAP is treated outpatient Sixth leading cause of death Inpatient mortality 10-35% One-year mortality = 40% (vs. 29% in controls) Outpatient mortality < 1% Some evidence that quality of care for African-Americans with CAP is worse Higher mortality among Caucasians Mortensen EM, et al. BMC Health Serv Res. 2004;4:20. Mayr FB, et al. Crit Care Med. 2010;38:759. 3
Cough 90%* Dyspnea 66% Sputum 66% Pleuritic chest pain 50% Less classic presentations 10% have NONE of the classic signs or symptoms Up to 40% will not have fever Up to 45% will have altered mental status * Yet, only 4% of all visits for cough are pneumonia Halm EA, Teirstein AS. N Engl J Med 2002;347(25):2039. Mehr DR, et al. J Fam Prac 2001;50(11):1101. Riquelme R, et al. Am J Respir Crit Care Med 1997;156:1908. Community-Acquired Pneumonia Background Etiology Diagnosis Treatment Prevention Typical organisms S. pneumoniae, H. influenzae, M. catarrhalis, etc. 4
Atypical organisms M. pneumoniae, C. pneumoniae, Legionella spp, etc. Classic teaching is not supported by the literature Some general trends S. pneumoniae in older pts, co-morbidities Mycoplasma in patients < 50 years old But - no history, exam, laboratory, or radiographic features predict organism Walking pneumonia Classic lobar pneumonia Outpatients (mild) Non-ICU inpatients ICU inpatient Outpatients (mild) Non-ICU inpatients ICU inpatient GNRs S aureus (40-50%) GNRs S aureus File TM. Lancet 2003;362:1991. Metlay JP, et al. JAMA 1997;278(17):1440. File TM. Lancet 2003;362:1991. 5
Outpatients (mild) Non-ICU inpatients ICU inpatient Outpatients (mild) Non-ICU inpatients ICU inpatient (40-50%) GNRs S aureus (40-50%) GNRs S aureus File TM. Lancet 2003;362:1991. File TM. Lancet 2003;362:1991. Outpatients (mild) (40-50%) Non-ICU inpatients ICU inpatient GNRs S aureus Community-acquired MRSA CAP First reported in 2003; increasing Specific strain of MRSA -- Panton-Valentine Leukocidin (PVL) genes File TM. Lancet 2003;362:1991. 6
Clinical Features % of Patients Flu Prodrome 30-75% Shock 50-100% Multi-lobar 50-100% Necrotizing 33-100% Leukopenia 25-100% Ventilated 50-100% Mortality ~ 40% Outpatients (mild) Non-ICU inpatients ICU inpatient GNRs S aureus (MRSA) Lancet. 2009. File TM. Lancet 2003;362:1991. Metlay JP, et al. JAMA 1997;278(17):1440. Background Etiology Diagnosis Treatment Prevention a. Treat for community-acquired pneumonia. b. Send him for a PA and lateral CXR. c. Send him for blood and sputum cultures. d. Prescribe sudafed and robitussin and send him home. e. Perform trans-tracheal aspiration f. B and C 7
1) Select clinical features (e.g. cough, fever, sputum, pleuritic chest pain) AND 2) Infiltrate by CXR or other imaging All expert guidelines state should have positive CXR to make diagnosis History, exam, etc. not good enough In outpt setting, should see an infiltrate. Order CXR if you are concerned about CAP If CXR negative, likely should no treat for CAP In the inpatient setting, can you see pneumonia with a negative CXR??? IDSA/ATS Guidelines. CID. 2007;44:S27-72. Community-Acquired Pneumonia Should order CXR in all patients with suspected pneumonia. In the hospital, a positive CXR is not necessary to treat as CAP (but consider other diagnoses). a. Treat for community-acquired pneumonia. b. Send him for a PA and lateral CXR. c. Send him for blood and sputum cultures. d. Prescribe sudafed and robitussin and send him home. e. Perform trans-tracheal aspiration f. B and C Community-Acquired Pneumonia 8
a. Treat for community-acquired pneumonia. b. Send him for a PA and lateral CXR. c. Send him for blood and sputum cultures. d. Prescribe sudafed and robitussin and send him home. e. Perform trans-tracheal aspiration f. B and C Provides a specific diagnosis The data: Cultures within 24 hours of arrival are associated with 10% lower odds of 30d mortality No evidence of benefit in outpatient setting Meehan TP, et al. JAMA 1997;278. Limitations Positive in < 10% of cases Medicare database of 13,000 patients Determined predictors of blood-culture positivity The sicker they are... Metersky ML, et al. Am J Respir Crit Care Med 2004;169 Metersky ML, et al. Am J Respir Crit Care (3):342-7 Med 2004;169(3):342-7 9
Limitations Positive in < 10% of cases High percentage of contaminants Cultures change antibiotics in < 5% of patients Costly In general, blood cultures are not indicated in the management of outpatient CAP. For inpatient CAP, blood cultures are optional unless clear risk factors for positive blood cultures: ICU, severe liver disease, cavitary infiltrates, pleural effusion Metersky ML, et al. Am J Respir Crit Care Med 2004;169 (3):342-7 IDSA/ATS Guidelines. CID. 2007;44:S27-72. Complicated and controversial Simple, inexpensive, specific for pneumococcus Problems include: Up to 30% could not produce adequate sputum Good quality available in only 14% Most don t narrow antibiotics In general, sputum cultures are not indicated in the management of outpatient CAP For inpatient CAP, sputum is indicated: High-quality specimen, right to the lab ICU, Cavitary infiltrates, Underlying lung disease IDSA/ATS Guidelines. CID. 2007;44:S27-72. 10
Get the CXR (esp. outpatient) Blood and sputum cultures generally discouraged in outpatient CAP General trend away from blood and sputum in non-icu patients admitted with CAP Pneumococcal urinary antigen Rapid test, specificity > 90% If positive, tx for pneumococcal disease May not reduce antibiotic spectrum -- > 70% with no change Sorde R, et al. Arch Intern Med. 2011;171:166. Procalcitonin: precursor of calcitonin No hormonal activity Inflammatory marker Increased in sepsis, bacterial infection Intl J. Lung Dz. 2006 11
Procalcitonin Based Management Swiss RCT: 1360 pts with an upper respiratory illness 93% admitted Usual care vs. PCT assay PCT validated algorithm for stopping antibiotics If PCT was below XX, antibiotics were not started or stopped Procalcitonin Based Management Control PCT Group Outcomes* 20.2% 16.1% Abx Given 90% 78% Abx duration 10.7 days 7.2 days Abx reaction 33.1% 23.5% * Defined as death, ICU admit, complications, recurrent infection" Schuetz, et al. JAMA. 2010;302:1059. Schuetz, et al. JAMA. 2010;302:1059. Background Etiology Diagnosis Background Etiology Diagnosis Treatment Prevention Treatment Prevention 12
A. Levofloxacin PO B. Azithromycin PO C. Ertapenem IV D. Augmentin PO and Azithromycin PO E. Doxycycline PO and penicillin PO Outpatients (mild) Non-ICU inpatients ICU inpatient Outpatients (mild) GNRs S aureus Must cover all these organisms File TM. Lancet 2003;362:1991. 13
Outpatients (mild) Penicillin, erythromycin, macrolides, etc. Wimpy pneumococcus Drug-resistant angry S. pneumoniae (DRSP) Age > 65 years old Chronic disease Heart, lung, renal, liver Diabetes mellitus Alcoholism Malignancy (active) Immunosuppression Antibiotics in the last 3 months Outpatients (mild) Wimpy pneumococcus Drug-resistant S. pneumoniae (DRSP) 14
Outpatient, healthy, no DRSP risk factors Doxycycline or macrolide Macrolide = azithro, clarithro, erythro Age > 65 years old Chronic disease Heart, lung, renal, liver Diabetes mellitus Alcoholism Malignancy Immunosuppression Antibiotics in the last 3 months 15
Outpatient, DRSP risk factors Oral fluoroquinolone OR Oral β-lactam + doxy or β- lactam + macrolide Outpatient, DRSP risk factors Oral fluoroquinolone OR Oral β-lactam + doxy or macrolide (DRSP = drug-resistant angry strep pneumo) Oral fluoroquinolone: moxi, gemi, levofloxacin NOTE: macrolides are no longer indicated for outpatients with DRSP risk factors (US resistance > 40%) β-lactam: High-dose amoxicillin (1mg PO tid) Augmentin (875mg PO bid) A. Levofloxacin PO B. Azithromycin PO C. Ertapenem D. Augmentin PO and Azithromycin PO E. Doxycycline PO and penicillin PO Age > 65 years old Chronic disease Heart, lung, renal, liver Diabetes mellitus Alcoholism Malignancy Immunosuppression Antibiotics in the last 3 months 16
Outpatient, healthy, no DRSP risk factors Doxycycline or macrolide Outpatient, DRSP risk factors Oral fluoroquinolone OR Oral β-lactam + doxy or β- lactam + macrolide Non-ICU inpatients ICU inpatient (resistant) GNRs S aureus 17
Inpatient, non-icu Fluoroquinolone OR β-lactam + macrolide* Inpatient, ICU IV β-lactam + macrolide + vancomycin OR IV β-lactam + fluoroquinolone + vancomycin ICU inpatient (resistant) GNRs MRSA * At UCSF, we use ceftriaxone & doxycycline Benefits Inpatient, non-icu Fluoroquinolone OR β-lactam + macrolide Inpatient, ICU IV β-lactam + macrolide + vancomycin OR IV β-lactam + fluoroquinolone + vancomycin Frei CR, et al. Am J Med. 2006;119:865 - Retrospective study of 631 pts with CAP - Early switch to orals, shorter LOS, lower mortality Mortensen EM, et al. Am J Med. 2006;119:859. - Retrospective study of 787 pts with CAP - Decreased mortality at 48 hours Mortensen EM, et al. Am J Med. 2004;117:726-31 - Improved 30-day mortality 18
Benefits Arnold FW, et al. Arch Intern Med. 2009;169:1515. - Retrospective study of 1725 pts with CAP; - Shorter LOS, mortality 10% lower (NNT = 10) Mccabe C, et al. Arch Intern Med. 2009;169:1525. - Retrospective study of > 50,000 with CAP - Decreased mortality by 30% Asadi L, et al. Respir Med. 2012;106:451. - Retrospective study of 2973 outpatients with CAP - Decreased mortality by 77% Risk of Clinical Failure Meta-analysis of 15 RCTs, 2796 patients with mild to moderate CAP Compared short-course (< 7 days) with longer courses. Looked at clinical failure, bacterial eradication, and mortality. Li JZ, et al. Am J Med. 2007;120:783. 19
Relative Risk of Mortality No difference in clinical failure No difference in bacterial eradication No difference in mortality In subgroup analysis, trend toward favorable efficacy with short-course. Li JZ, et al. Am J Med. 2007;120:783. Patients with CAP should be treated for a minimum of 5 days (level I evidence) -- IDSA/ATS Guidelines Minimum of 5 days If afebrile for 48-72 For most, 7 days total 20
Have pts take temp q8 hours & report if > 101 o F or if > 99 o F after 48 hours Encourage pts to drink 1-2 quarts of liquid daily If they respond appropriately, have them follow-up within 10-14 days Standard practice? Prior ATS guidelines said yes, recent guidelines do not address CXR resolution: At 28 days, ~ 50% had not resolved Can consider in high-risk patients Significant smoking history, etc. Probably should wait > 3 months Niederman M. Ann Intern Med. 2009 Bruns AH. CID. 2007;45:983. Background Etiology Diagnosis Treatment Prevention Vaccine against Streptococcus pneumoniae, most common cause of CAP Polysaccharide vaccine with 23 antigens Covers 85-95% of serotypes causing invasive disease in U.S. ACIP Recs. MMWR 1997;46:1. 21
Updated meta-analysis of 22 trials, > 100,000 patients Only high-quality, blinded studies Relative Risk All-cause mortality 0.97 (0.87-1.09) All-cause pneumonia 1.19 (0.95-1.49) So, it doesn t prevent CAP or decrease mortality from CAP. Pneumovax may prevent invasive disease. Should we be giving it at all? ** No difference for elderly or chronic illness Huss A, et al. CMAJ. 2009;180:48. Four different trials looking at benefits of pneumovax in patients hospitalized with CAP. Compared vaccinated vs. non-vaccinated Looked at impact on ICU admission, inpatient mortality, inpatient complications, and LOS. Variable ICU admission Inpt complications LOS Inpt mortality Outcome Decreased Decreased Decreased Decreased 22
Pneumococcal vaccine likely prevents invasive pneumococcal disease. Probably reduces death, ICU admission, complications, and LOS in patients hospitalized with CAP. Quality Improvement Adults aged < 65 years Prevents influenza illness in ~ 70-90% Adults aged > 65 years Prevents influenza illness in ~ 30-70% Hospitalization Risk Reduction Resp. Illness 56%* Hospitalization 50%* Pneumonia 53%* All cause death 68%* (NNT = 118) * All p values < 0.001 ACIP Recs. MMWR 2003;52:1. Gross PA, et al. Ann Intern Med. 1995;123:518-27. 23
Hospitalization Risk Reduction Hospitalization 27%* for pna/flu All cause death 48%* Should provide it to all of our patients that smoke Some evidence that tobacco is a risk factor for pneumonia * All p values < 0.001 Nichol KL, et al. N Engl J Med 2007;357:1373. (Oct 4, 2007) Smoking and Invasive Pneumococcus Cigs / d OR Current Smokers 1-14 2.3 15-24 3.7 25 5.5 All 4.1 (2.4-7.3) Hrs / d OR Passive Smokers 4 2.4 > 4 3.9 All 2.5 (1.2-5.1) Background Etiology Diagnosis Treatment Prevention No PPI for you Nuorti, NEJM, 2000 24
Gulmez, et al. Arch Intern Med. 2007. -- Current use of PPI: CAP OR = 1.5 -- Recent start: CAP OR = 5.0 Sarkar, et al. Ann Intern Med. 2008. -- Recent PPI start: CAP OR = 3.8 Herzig, et al. JAMA. 2009. -- 52% of hosp pts got PPI, HAP OR = 1.3 Eurich, et al. Am J Med. 2010. -- Rates recurrent CAP after CAP admit -- Starting PPI: OR 2.1% (7% abs risk) Knol W, et al. JAGS. 2009. -- Recent anti-psychotic start (1 wk); OR 4.3** Trifiro, et al. Ann Intern Med. 2010. -- Population based study, 2000 patients. Current Use Risk of pneumonia Typical anti-psychotic OR = 2.6 (1.4-4.6) Atypical OR = 1.8 (1.2-5.3) Background Etiology Diagnosis Treatment Prevention Etiology: No predictors for typical or atypical need to treat both Etiology: Recognize CA-MRSA as a cause for severe CAP Diagnosis: Do not routinely get blood or sputum cultures in outpt CAP 25
Treatment: doxycycline or macrolide for healthy outpatient with no DRSP risk-factors Treatment: fluoroquinolone or β-lactam + macrolide/doxy for outpt with DRSP risk factors Treatment: most 7 days Prevention: pneumovax, flu vax, smoking, avoid PPIs Bradley A. Sharpe, M.D. Associate Professor Medicine Department of Medicine UCSF sharpeb@medicine.ucsf.edu Validated severity-of-illness scoring system Retrospective then prospective Advocated by the British Thoracic Society Based on five easily measurable clinical factors Lim WS. Thorax 2003; 58:377 82 26
CURB65 Score Mortality 0 0.7% 1 2.5% 2 11.0% 3 15.5% 4 40.6% 5 57.0% Lim WS. Thorax 2003; 58:377 82 Lim WS. Thorax 2003; 58:377 82 CURB65 Mortality Treatment 0 0.7% Outpatient 1 2.3% Outpatient 2 11.0% Inpatient 3 15.5% Inpatient 4 40.6% ICU 5 57.0% ICU Consider using a prognostic score in CAP Especially useful in borderline cases CURB65 easy to use Lim WS. Thorax 2003; 58:377 82 27