1. Nosocomial Pneumonias
IDSA/ATS 2016 Guidelines
José Luis González, MD
Clinical Assistant Professor of Medicine

2. Outline Intro - Definitions & Diagnosing VAP & HAP treatment
De-escalating empiric therapy

3. IDSA/ATS 2016 Guidelines Management of Adults with Hospital-acquired and Ventilator-associated Pneumonia
Use of Gradings of Recommendations Assessment, Development & Evaluation (GRADE)
No more HCAP
Encouraging development of hospital-specific antibiograms
Replaces 2005 guidelines

4. Definitions No gold standard
Pneumonia: new lung infiltrate w/ clinical evidence of infectious origin:
New onset fever
Purulent sputum
Leukocytosis
Decline in oxygenation

5. Definitions Healthcare-Associated Pneumonia HAP: VAP:
pna not incubating at time of hospital admission and
occurring 48° post admission
VAP:
pna occurring 48° post endotracheal intubation

6. Sputum Samples Non-invasive sampling > invasive sampling (wk, lq)
Hold abx if Cxs result < threshold CFUs (BALs (wk, vlq)
Tx based on microbiologic evidence rather than empirically (wk, vlq)
Only 30% of resp infections due to MRSA in pts w/ +MRSA screen
Meta-analysis: poor concordance between gram stains and final cultures
Invasive techniques: BAL, mini-BAL (blind bronchial sampling) and PSB (protected specimen brush)
Non-invasive techniques: spontaneous expectoration, sputum induction, nasotracheal suctioning, endotracheal aspirate if intubated
PSB < 103 CFUs
BAL < 104 CFUs
ETA < 106 CFUs
(Semi-quant: light, moderate or heavy) vs (104 , 105 or 106)
There is an association between +MRSA screens and MRSA infection, but usually only in skin and soft-tissue infections

7. LAC-USC Antibiogram

8. Ventilator-Associated Pneumonia
Empiric Treatment

9. Organisms Gram Negative Bacilli Staphylococcus Aureus: MRSA vs MSSA
Pseudomonas Aeruginosa: single agent vs double-coverage

10. VAP: Gram negative bacilli & MSSA
Cefepime 2g IV q8h or
Piperacillin-Tazobactam 4.5g IV q6h
Levofloxacin 750 IV qDay / Ciprofloxacin 400mg IV q8h
Meropenem 1g IV q8h / Imipenem 500mg IV q6h
If sensitive to polymixins, use IV polymyxin (colistin or polyB) and adjunctive inhaled colistin
Aminoglycosides have poor lung penetration so avoid: gentamycin or tobramycin
- tigecycline and doripenem were associated w/ worse outcomes

11. VAP: When to treat for MRSA (sr, lq)
IV abx within 90 days prior
>10-20% or unknown MRSA rate LAC)
Renal replacement therapy
Hospitalized 5d prior to intubation
Septic time of intubation
ARDS preceding intubation
LAC 34% MRSA
Septic shock: change in SOFA > 2 + pressors + lactate > 2

12. Treating for MRSA Vancomycin 15 mg/kg IV q8-12h or
Linezolid 600 mg IV q12h
Avoid aminoglycosides due to poor lung penetration and increased risk of nephrotoxicity and ototoxicity
Avoid colistin

13. VAP: When to treat with 2 antibiotics for pseudomonas (wr, lq)
IV abx within 90 days prior
>10% resistance rates to single agent being considered LAC)
Renal replacement therapy
Hospitalized 5d prior to intubation
Septic time of intubation
ARDS preceding intubation
No diff in mortality, clinical response, adverse effects or acquired resistance rates between regimens with one antipseudomonal agent vs 2. but data do not apply to all pts bcz most studies excluded pts colonized with resistant pathogens and patients at increased risk for resistant pathogens.
Lower threshold for dbl coverage for pseudomonas than for MRSA because gram negs are more frequently implicated in VAP = increased rate of inadequate coverage

14. Agents with activity against pseudomonas
Cefepime 2g IV q8h or
Piperacillin-Tazobactam 4.5g IV q6h
Levofloxacin 750 IV qDay / Ciprofloxacin 400mg IV q8h
Meropenem 1g IV q8h / Imipenem 500mg IV q6h

15. 100
100
34%
LAC-USC Antibiogram 87 87 81 76 87

16. Hospital-Acquired Pneumonia
Empiric Treatment

17. HAP: Gram negative bacilli & MSSA
Cefepime 2g IV q8h or
Piperacillin-Tazobactam 4.5g IV q6h
Levofloxacin 750 IV qDay / Ciprofloxacin 400mg IV q8h
Meropenem 1g IV q8h / Imipenem 500mg IV q6h
If sensitive to polymixins, use IV polymyxin (colistin or polyB) and adjunctive inhaled colistin
Aminoglycosides have poor lung penetration so avoid: gentamycin or tobramycin
- tigecycline and doripenem were associated w/ worse outcomes

18. HAP: When to treat for MRSA (sr, vlq)
IV abx within 90 days prior
>20% or unknown MRSA rate LAC)
Septic shock
Need for vent support due to pna
10% of HAP is MRSA

19. Treating for MRSA Vancomycin 15 mg/kg IV q8-12h or
Linezolid 600 mg IV q12h

20. HAP: When to treat with 2 antibiotics for pseudomonas (sr, vlq)
IV abx within 90 days prior
Septic shock
Need for vent support due to pna
Structural lung disease (bronchiectasis or CF)
High qual. Gram stain w/ numerous predominant GNB
Pseudomonas is the cause of about 13% of HAPs
Other GN orgs = 22%
Overall = 35%

21. Agents with activity against pseudomonas
Cefepime 2g IV q8h or
Piperacillin-Tazobactam 4.5g IV q6h
Levofloxacin 750 IV qDay / Ciprofloxacin 400mg IV q8h
Meropenem 1g IV q8h / Imipenem 500mg IV q6h

22. VAP HAP Pseudomonas MRSA IV abx within 90 days prior
>10-20% or unknown MRSA rate
Renal replacement therapy
Hospitalized 5d prior to intubation
Septic time of intubation
ARDS preceding intubation
>10% resistance rate to single agent being considered
IV abx within 90 days prior
>20% or unknown MRSA rate
Septic shock
Need for vent support due to pna
Structural lung dz (bronchiectasis or CF)
Gram stain w/ +++ predominant GNB
Pseudomonas MRSA
HAP pts tend to be less sick than VAP pts, thus pt w/ HAP and pseudomonal pna (13% of cases) means that if you guess wrong and don’t double cover, these pts are less sick and no evidence that this increases mortality

23. Use Pharmakokinetic Data for Dosing (wk, vlq)
3 RCTs and 4 observational trials:
improved clinical cure rate
Reduced mortality
Reduced length of ICU stay
Esp due to comorbidities of patients in the ICU

24. Duration of treatment VAP: 7 days (st, mq) HAP: 7 days (st, vlq)
1 meta-analysis 508pts (7-8d) showed no diff in:
Mortality, recurrent pna, duration of vent, hosp stay
Another 883pts (7-8d vs 10-15d) showed no diff in:
Mortality, recurrent pna, duration of vent, length of ICU stay
Subgroup of pseudomonas & Acinetobacter = slight increase in recurrence
HAP: 7 days (st, vlq)
No studies on HAP duration of therapy; recs based on VAP
Increased recurrence in subgroup analysis (OR, 2.18; 95% CI )
Authors own meta-analysis including the two studies above showed no diff in subgroup analysis
recommendation was not altered for these pt bcz mortality and clinical cure were not affected

25. Empiric vs Targeted therapy
Targeted > Empiric (st, lq)
For pts w/ confirmed pseudomonas w/ known susceptibilities; ok to use monotherapy unless pt is: (wk, vlq)
In septic shock
At high risk of death
No def of high risk of death: previously mentioned necessitating vent support
High risk > 25%
Low risk < 15%
Consider pts w/ high SOFA score
The exception is based on (a single) observational trial(s)
For Acinetobacter species: tx w/ either carbapenem or ampicillin/sublactam

26. HAP & VAP: Should therapy be de-escalated? (wk, vlq)
6 trials: 1 RCTs, 5 observational trials
When pooled, no diff in mortality
2nd RCT: some worse outcomes but pna subgroup = no diff
Reasonable to de-escalate to single broad spectrum agent in pts who:
Have negative cultures
Are clinically improving
2nd RCT showed increase antimicrobial days and increased risk of superinfection, but in a subgroup of the pna patients, no difference was found

27. Ventilator Associated Tracheobronchitis
Ventilated patient with:
Fever
New/increased sputum production
+ endotrach aspirate cx (ETA ≥ 106 CFUs)
And no radiographic evidence of pneumonia
Do not treat (wk, lq)

28. Biomarkers: Procalcitonin (PCT)
Initiation
LAC says: yes
IDSA/ATS says: no (st, mq)
 no studies comparing patient outcomes
6 studies w/ significant sources of bias: 67% sens, 83% spec
Discontinuation
Meta-analysis of 14 trials: no diff in mortality or tx failure
3 other RCTs: shorter duration of abx tx (9 vs 12d) w/o diff in outcomes
Endotoxin mediated release stimulated by K-cells in the lung
Also stimulated from inflammation or viral illness, but to a lesser extent
6 studies w/ 665 pts, 50% ultimately diagnosed w/ HAP/VAP
AUROC = 0.76
most studies showed positive predictive value however
varying methods for assay
different cut-offs
pts enrolled did not have legitimate diagnostic uncertainty
A-priori cutoff: 90% sens and spec
if 1000pts, this cutoff would result in 5% (50pts) pts incorrectly diagnosed w/ a pna and 5% of pts incorrectly diagnosed as not having one
with current cutoffs: 8.5% (85pts) incorrectly with one and 16.5% (165pts) = incorrectly diagnosed w no pna
Other outcomes measured in DC trials: vent, length of ICU stay, recurrence or resistance

29. Procalcitonin Algorithm @ LAC-USC
8.5% (85pts) incorrectly with a PNA and 16.5% (165pts) = incorrectly diagnosed without a PNA
In pts at high likelihood for infection, I would NOT use procalcitonin. Also, if pt remains septic, I would not de-escalate therapy as per IDSA/ATS guidelines
Finally, studies decreased abx use from 12d to 9 days. Guidelines recommend only tx x 7days
If using this guide, I would omit pts w/ a diagnosis of pneumonia. (although most of the data on PCT is in PNAs)
Procalcitonin Algorithm @ LAC-USC

30. Biomarkers Don’t use procalcitonin (PCT) to initiate therapy (st, mq)
PCT better than clinical criteria alone for de-escalation (wk, vlq)
Don’t use sTREM-1, CRP, CPIS to guide therapy
Soluble triggering receptor expressed on myeloid cells: TREm-1 is a member of the Ig family and has been shown to be strongly expressed on neutrophils and monocytes infiltrating tissues invaded by bacteria for fungi. Recent studies show that it is elevated in non-infectious causes of inflammation too
CPIS Modified clinical pulmonary infection score: temp, wbc, character of resp secretions, PaO2/FiO2 ratio and CXR findings

31. Summary HAP & VAP occur 2 days post inciting event
Obtain non-invasive cultures
Use antibiogram
Use pharmacokinetic data
De-escalate if pt improves, once sensitivities result. Ok to use PCT

32. References
Kalil AC, Metersky ML, et al. Management of adults with hospital- acquired and ventilator-associated pneumonia: 2016 clinical practice guidelines by the Infectious Diseases Society of America and the American Thoracic Society. Clin Infect Dis. 2016;63(5):e61–111.
LAC + USC Antibiogram; accessed 12/5/2016
Holtom P, Spellberg B. PCT Algorithm for Initiating and Stopping Antibiotics in Adult Patients with Sepsis in the ICU; accessed 12/5/2016