Disclosures I have nothing to disclose Updates In CAP/HAP/VAP - - PDF document
3/20/2019 Disclosures I have nothing to disclose Updates In CAP/HAP/VAP Rachel Bystritsky, MD University of California, San Francisco Outline Outline CAP: CAP Epidemiology - Epidemiology Microbiology - Microbiology
I have nothing to disclose
CAP:
Epidemiology Microbiology Diagnosis Management
HAP/VAP
Diagnosis Management
HCAP
Talk will focus on adults
Excluding severely immunocompromised pts
2016 CDC data
48,632 deaths1 15.1 deaths/100,000 population1 PNA+Influenza 8th leading cause of death in US2 California: 5,981 deaths from Influenza+PNA3
Most common cause of death from infectious disease
ED visits: 544,000 in 2015
1.
National Vital Statistics Reports, Vol. 67, No. 5, July 26, 2018
2.
National Vital Statistics Report, Vol. 67, No 6, July 26, 2018
3.
CDC National Center for Health Statistics
4.
National Hospital Ambulatory Medical Care Survey: 2015
CID 2017; 65: 1806-1812
CID 2017; 65: 1806-1812
No etiological agent is found in the majority of cases even using advanced techniques
S pneumoniae is the most commonly detected bacterial pathogen
Respiratory viruses increasingly recognized
Role remains unclear: single pathogen, co-pathogen
Jain S et al. N Engl J Med 2015;373:415-427.
CDC: VPD Surveillance Manual Pneumococcal: Chapter 11.1
On rates of invasive pneumococcal disease in adults
Antibiotic Susceptible (%) Intermediate (%) Resistant (%) Penicillin 95.7 2.0 2.4 Cefotaxime 97.6 2.0 0.4 Erythromycin 69.6 0.2 30.1 Tetracycline 88.9 0.2 10.9 Levofloxacin 99.9 0.0 0.1 Vancomycin 100.0
Invasive Pneumococcal Isolates, 20151
Trends in Pneumococcal Resistance
CDC: Active Bacterial Core surveillance
Red = Erythromycin Green = Cefotaxime
CXR should be obtained in pts with suspected PNA
History and exam have inadequate spec/sens for dx
Identifies complications (i.e. empyema)
CT: Has higher sensitivity and specificity than CXR3
Higher exposure to radiation and cost No evidence re: effect on outcomes Should be reserved for high risk patients with
inconclusive CXR
Ultrasound: may be a helpful adjunctive modality4
Requires an experienced operator
1.
Arch Intern Med 1999; 159: 1082-1087.
2.
JAMA 1997; 278: 1440-1445.
3.
Am J Respir Crit Care Med 2015; 192: 974-982.
4.
Chest 2017; 151: 374-382.
Sputum culture:
Helpful if single pathogen is detected – can guide de-
escalation or change in abx if poor response/adverse rx to empirical therapy
Many patients can not produce an adequate sample1 Sensitivity is poor, worse if abx given prior1 Should be obtained in hospitalized patients
Blood cultures:
Positive 5-16% in hospitalized patients1,2 Higher yield in higher severity illness
1. Eur J Clin Microbien Infect Dis 2006; 24:241-249. 2. Chest 1995; 108: 932.
IDSA/ATS Criteria have poor sens/spec for predicting which patients will have a positive test1
Strep Pneumo (SP): 61% sens, 39% spec Legionella (serotype 1) (LP): 63% sens, 35% spec
No clinical features useful for predicting +SP test
Clinical features associated with + LP test:
Hyponatremia, fever, diarrhea, and recent travel
Influenza PCR when virus circulating
Respiratory viral PCR
Can potentially help with de-escalation Caution in interpretation
may be co-pathogen or incidental finding
Bronchoscopy
Consider for non-response to tx or concern for
unusual pathogen
Precursor of calcitonin, rises in response to bacterial infection
Levels are higher in bacterial than viral infections
Can be used as an adjunct to help guide initiation
Pts with low PCT (< 0.25 used at UCSF) less likely to
have bacterial infection (can consider not starting therapy in low risk patients)
Discontinuation of therapy once PCT less than 0.25 or
decreased ≥80% from peak
Large RCT in hospitalized patients showed no impact on antibiotic use1
Limitations:
No threshold value perfectly distinguishes viral vs
bacterial infection
Often not elevated with intracellular pathogens (i.e.
Legionella and Mycoplasma)1
Not validated in special populations (surgery,
pregnancy, immunocompromised)
Should not replace clinical judgment
Once dx is made, next step is determining severity
Outpatient vs. inpatient
Validated CAP severity scores:
PORT Score CURB-65 SMART-COP
CURB-65 is the simplest but may not be as sensitive as PORT Score or SMART-COP1
PORT Score best validated, identifies patients that can be safely managed as outpatients
Stratifies patients into 5 classes
Includes 20 variables:
Demographics Comorbidities Exam findings Laboratory findings (including blood gas) Imaging (presence/absence of pleural effusion)
N Engl J Med 1997;336:243-250.
N Engl J Med 1997;336:243-250.
A)
Yes
B)
No
Oversimplifies interpretation of predictor variables:
Pt in previous question would have a port score of 45
(Class II: outpatient treatment reasonable)
Developed inn cohort that excluded immunocompromised, pregnant patients
Does not take into consideration important co- factors (i.e. psychosocial factors, ability to take PO
Simplest scoring system:
Confusion BUN Respiratory rate Systolic BP Age
Sensitivity for predicting ICU admission/critical care intervention is poor1:
Of Pts with scores of 0 or 1 (rec outpatient
management) 15.6% were admitted to ICU and 6.4% received critical care intervention
Pt in question would have a score of 1: consider
A)
Amoxicillin
B)
Azithromycin
C)
Levofloxacin
D)
Doxycycline
E)
Amoxicillin + azithromycin
Outpatient (empirical):
Previously healthy no risk factors for drug resistance
A macrolide (1st choice) Doxycycline
Comorbidities, prev abx within 3 mo, or high rate of
drug resistant Strep pneumo
Respiratory fluoroquinolone β-lactam + macrolide (1st choice) or doxycycline
Recommend avoiding macrolide monotherapy if local rate of resistance >25% for S pneumo
Inpatient (empirical): non-ICU
Respiratory fluoroquinolone β-lactam + macrolide
Inpatient (empirical): ICU
β-lactam + macrolide or fluoroquinolone
β-lactam = ceftriaxone, cefotaxime or amp-sulbactam
This guideline has been archived and is currently being updated: projected publication Fall 2019
Fluoroquinolones: FDA warnings
2008: tendonitis and tendon rupture 2013: may cause peripheral neuropathy 2016: increased risk of disabling and potentially
permanent side effects involving tendons, muscles, joints, nerves, and the central nervous system
Reserve for use if no other options for bacterial
sinusitis, acute bacterial exacerbation of chronic bronchitis or urinary tract infection
July 2018: strengthened warning re CNS effects and
hypoglycemia
Dec 2018: increased risk of aortic dissection
Azithromycin for Pneumonia
2012 Tennessee Cohort Study1:
Compared pts rx’d azithromycin compared to no
antibiotics, amoxicillin or quinolone
Increased risk of CV death and death from any cause Increased risk compared to no abx, amox, or cipro No significant difference compared to levoflox
2013 FDA “safety communication” re: increased risk
2014 VA Cohort Study2: hospitalized patients
Increased risk of MI but decreased risk of mortality with
azithromycin
Excluded patients who received doxycycline
1. NEJM 2012; 366: 1881-1890 2. JAMA 2014; 311: 2199-2208
Is atypical coverage necessary for all patients?
British guidelines1,2 recommend amoxicillin alone as
first line agent for low severity PNA (vs doxycycline
reserved for moderate or high severity PNA
S pneumo rates of macrolide resistance are >25% Studies investigating benefits of atypical coverage for
hospitalized patients are mixed
1. 2009 BTS Guidelines 2. 2014 NICE Guidelines
Cluster randomized trial of hospitalized patients admitted to non-ICU wards
Compared β-lactam monotherapy to β-lactam (656 pts) + macrolide and β-lactam (739 pts)+ fluoroquinolone
Allowed deviation from protocol for medical reasons
92.7% of monotherapy pts treated per protocol
Found β-lactam monotherapy non-inferior to combination therapy for 90 day mortality (and LOS, complications)
Limitations:
Conducted in Netherlands – limits generalizability 90 day mortality not the best outcome for low risk
patients
NEJM 2015; 372:1312-1323
Numerous retrospective observational studies have shown mortality benefit of combination therapy for critically ill patients
Meta-analysis of 28 observational studies with almost 10,000 studies showed relative risk reduction of 18% (3% absolute risk reduction) for combination therapy1
No prospective clinical trials
Should rem ain standard of care for severe CAP
Has broad activity against bacterial causes of CAP:
S pneumo, H flu, Moraxella, Mycoplasma, C
pneumoniae, some activity against Legionella
Local susceptibility data for S pneumo:
UCSF 2017: 94% susceptible SFGH 2016: 79% susceptible
Better tolerated than macrolides and fluoroquinolones
Clinical data is limited but widely used
For hospitalized patients receiving ceftriaxone, receipt of doxycycline is associated with a lower risk of C difficile infection
Retrospective cohort at SFGH 2005-2010 1066 pts who received doxycycline vs. 1668 who did
not
CDI incidence was 8.11/10,000 patient days for
patients who did not receive doxycycline and 1.67/10,000 patient days for patients who did
Clin Infec Dis 2012; 55: 615-620
EPIC study – 2% of patients with CAP had S aureus PNA
Consider MRSA coverage (vanco or linezolid) for:
Patients with septic shock or requiring mechanical
ventilation
GPCs in clusters on gram stain Recent influenza like illness Necrotizing or cavitary pneumonia Emypema
Controversial
Studies have used highly variable regimens
Recent meta-analysis (2018): 6 studies, 1509 pts1
Adjunctive corticosteroids associated with:
Reduced time to clinical stability and length of stay (1d) No effect on mortality Increased risk of CAP-related readmission and
hyperglycemia
Trend towards increased mortality for low severity CAP
and trend towards decreased mortality for high severity
Heterogeneous patient population
Other meta-analysis have shown decrease in mortality for severe CAP2
1. Clin Infect Dis 2018: 66:346-353. 2. Ann Intern Med 2015;163(7):519-28
Pending studies for severe CAP
ESCAPe Trial: methylprednisolone
VA study, RCT, critically ill patients Completed, not yet published Enrolled 586 patients
CAPE_COD Trial: hydrocortisone
Multicenter study in France Not yet completed Goal enrollment 1200 patients
2007 IDSA/ATS guidelines:
minimum of 5 days (level I evidence), pts should be
afebrile for 48–72 h, and should have no more than 1 CAP-associated sign of clinical instability
Duration of 5-7 days is adequate for most patients
Multiple meta-analyses have shown similar efficacy of
short (≦7 days) vs. long courses of abx1,2,3
Procalcitonin guidance helps shorten therapy in settings where longer courses are routinely being used
Can switch to oral when the patient is stable and tolerating PO
1. Am J Med. 2007;120(9):783-90 2. Drugs 2008; 68: 1841-1851 3. Intern Emerg Med 2018. Epub ahead of print
Most patients are treated for too long
2012-2013 data: Median length of therapy 9.5 days1
When should a longer duration be used?
Necrotizing pneumonia/lung abscess Empyema S aureus or Pseudomonas Cystic fibrosis/bronchiectasis
Drugs approved for CAP since 2007 guidelines:
Tigecycline –approved in 2009
High incidence of adverse reactions (GI) Associated with increased risk of death Reserved for MDR infections with limited options
Ceftaroline – approved in 2010
Covers MRSA but not approved for this indication
(limited evidence)
Less well tolerated than ceftriaxone
Omadacycline – approved 2019
Use for CAP will be limited Actual indication will likely be for MDR gram
negatives
Latest guidelines:
HAP: pneumonia that develops >48 hrs after hospitalization
VAP: pneumonia that develops >48 hrs after intubation
Pathogen % of Total S aureus 24.1% P aeruginosa 16.6% Klebsiella spp 10.1% Enterobacter 8.6% Acinetobacter 6.6%
Top 5 pathogens in VAP: 2009 – 2010 NHSN Data
ICHE 2013; 34: 1‐14
Respiratory cultures:
Try to obtain in all patients Non-invasive sampling preferred
HAP: expectorated, induced, nasotracheal suction VAP: tracheal aspirate
Use of procalcitonin is not recommended
Empirical treatment should depend on local antibiogram
Cover S aureus, Pseudomonas, other gram negative bacilli
Therapy should be modified based on culture results
Recommended if:
Risk factor for resistant organisms
VAP:
HAP+ VAP:
In unit with >10-20% of S aureus is MRSA* Options
* This includes most hospitals
Clinical utility of MRSA nares swab1:
Meta-analysis of 22 studies with 5163 patients
Included both culture and PCR Included CAP/HAP/VAP Pooled prevalence of MRSA pneumonia: 10% Sensitivity 70.9%, Specificity 90.3% for MRSA PNA PPV: 44.8%, NPV: 96.5%
Use of nasal MRSA PCR testing reduces duration of
empirical MRSA targeted therapy without increasing adverse clinical outcomes2
1. CID 2018; 67: 1-7 2. Antimicrob Agents Chemother 2017; 61: e02432-16
UCSF has implemented MRSA nares swab guided empirical therapy for HAP/VAP:
No MRSA coverage given if negative PCR within 7
days of onset
2016 IDSA/ATS guidelines:
2 antipseudomonal antibiotics should be used only if:
For VAP: Pt in unit where >10% of GN isolates resistant
to agent being considered for monotherapy or Pt high risk for resistance (Same RF’s as MRSA)
For HAP: if high risk for mortality or prior IV antibiotics
within 90 days
Avoid using aminoglycoside if alternative agent with
adequate GN coverage is available (i.e. FQ)
Do not use aminoglycoside as monotherapy
Antipseudomonal/Anti-MSSA agent:
Pip-tazo, cefepime, levofloxacin, imipenem or
meropenem
PLUS MRSA agent if RF or >10-20% of Staph is MRSA (>20% for HAP)
Linezolid or vancomycin
PLUS 2nd antipseudomonal agent if RF for resistance or in unit >10% resistance of GNs to primary agent (VAP) or high risk for mortality (HAP)
Duration: 7 days
Introduced in 2005 guidelines
HCAP Risk factors:
hospitalization for more than 48 hours in the last 90
days
residence in a nursing home or extended care facility home infusion therapy chronic dialysis within one month home wound care a family member with a multi-drug resistant organism.
Removed from 2016 guidelines
2014 Meta-analysis1:
24 studies (22,456 patients) “HCAP” associated with increased risk of MRSA and
Pseudomonas – however discriminatory power was poor
Studies were low quality Confounded by publication bias After adjustment for age and co-morbidities,
mortality not increased
? May be addressed in upcoming CAP guidelines
Most patients with “HCAP” by old guideline criteria can be treated as CAP
Consider expanded therapy if:
Severely ill History of resistant organisms Extensive antibiotic exposure
If using expanded therapy, prioritize microbiological dx and de-escalate based on results