Transcript Jim Hoehns, Pharm.D.

Healthcare –Associated Pneumonia

93 female LTCF patient presents with dyspnea
and fever.
 EMS notified; brought to ER



Medical hx: Afib, dementia, spinal stenosis, R
hip OA, LVEF 55%
Vitals: 137/97 P-98, 102F, RR 22, O2 initially 75%,
then 93% on 3L
Exam:
 Decreased breath sounds on the right
Labs: WBC 8.8, Cr = 0.7, Hgb = 14
CXR: congestive changes; right basilar patchy
opacity with positive infiltrate
 Diagnosed with pneumonia and admitted



93 y.o. F from LTCF with pneumonia.
Treatment?

A.
B.
C.
D.
E.




Levofloxacin
Levofloxacin + vancomycin
Levofloxacin + Pip/Tazo + vancomycin
Ceftriaxone
Ceftriaxone + azithromycin




P: Elderly LTCF patients with pneumonia
I: Guideline based antibiotics
C: Non-guideline based antibiotics
O: Mortality or clinical outcomes

HAP: hospital acquired pneumonia
 Arises 48hrs or more after admission

HCAP: health-care associated pneumonia**
 Hospitalized within last 90 days; LTCF, IV therapy,
chemotherapy, wound therapy, or attended a hospital
or hemodialysis clinic in last 30 days

VAP: ventilator associated pneumonia
 Arises more than 48-72 hrs after intubation

HCAP included in spectrum of HAP and VAP
 HCAP: need therapy for MDR pathogens


Acknowledges most evidence for VAP
Timing of pneumonia is important
 “Early onset HAP/VAP”: within 4 days of
hospitalization (likely sensitive bacteria)
 “Late onset HAP/VAP”: at 5 days or later (MDR
pathogens more likely)

HCAP etiology?
 “elderly residents of LTCFs have a spectrum of
pathogens that more closely resemble late-onset
HAP and VAP”
 Study 1
▪ Staph aureus (29%), enteric GNRs (15%), Strep
pneumoniae (9%), Pseudomonas (4%)
 Study 2 – “failed to respond to 72 hrs of abx”
▪ MRSA (33%), GNRs (24%), Pseudomonas (14%)
Goal: evaluate effectiveness of guideline-based therapy (GBT) compared with other
antimicrobial regimens and to identify subgroups of patients with HCAP who received
greatest benefit from GBT.


Cohort study; 346 U.S. hospitals
Inclusion
 Patient discharge between Jul 2007 – Jun 2010
 Age ≥ 18 yrs with
▪ Primary ICD-9 dx of pneumonia, OR
▪ Secondary dx of pneumonia, paired with primary dx of
respiratory failure, ARDS, respiratory arrest, sepsis, or
influenza

HCAP
▪ If dx of ESRD/dialysis in first 2 hospital days, OR if admit
from a SNF, OR if DC from hospital in past 90 days, OR
taking immunosuppressant drugs

Exclusion
 Transfer patients (could not assess initial severity





or outcomes)
Length of stay ≤1 day
Cystic fibrosis
Attending not expected to treat pneumonia
DRG inconsistent with pneumonia
Any pt who did not have a CXR and begin
antimicrobials within 48hrs of admission

Data elements
 Age, sex, race, marital and insurance status,
comorbidities, tests, medications and treatments,
physician specialty, comorbidities (via a software
program)
 Hospitals: region, bed size, rural/urban, teaching
status

GBT
 1 abx against MRSA and 1 abx against Pseudomonas
 Main predictor variable



Non-GBT: all other abx regimens
Primary outcome: in-hospital mortality
Secondary outcomes
 7 day mortality, initiation of mechanical ventilation or
admission to ICU, readmission in 30 days, cost,
length of stay, clostridium difficile infection (CDI)

Categorical variables
 Frequencies and proportions

Continuous variables
 Medians with IQRs

Logistic regression model for treatment
 Propensity scores for GBT vs. non-GBT


Adjusted analyses
Sensitivity analysis
 Explore effect of hypothetical unmeasured
confounders

N=85,097 patients from 346 hospitals
 31,949 (37.5%) received GBT
 Of those not receiving GBT, 82% received
standard therapy for CAP

GBT patients
 Younger
 More likely male
 More chronic disease
 More severe pneumonia
Sensitivity Analysis: “A single potential confounder would
have to be present in 30% of the GBT patients (and none
of the non-GBT patients) and have an OR of 3.0 in order
to find a statistically significant benefit to GBT”

Explanations
 Selection bias: physicians refer GBT for sickest
patients
 GBT might harm some patients
▪ ADE’s, resistance, CDI, complications of IV or prolonged
hospitalization




Observational nature
Worked solely with claims; could miss
important confounders
No microbiologic data
Modified the ATS/IDSA guidelines
 Used 1 vs. 2 drugs for pseudomonas
 (results were same regardless)




To date, no RCTs of GBT for HCAP
Findings question the necessity of treating all
HCAP patients with GBT
Better models needed to identify at-risk
patients for MDR pathogens
2010: Only 40% of patients with HCAP
receive GBT

Ewig S et al. Curr Opin Infect Dis 2012;25:166175.
 HCAP poorly predictive of MDR pathogens
 Frequency of MDR pathogens far lower than
supposed in the original guideline document
 HCAP concept results in tremendous overtreatment
without any evidence for improved outcomes


IDSA is currently revising HAP, VAP, and
HCAP guidelines
1 MRSA antibiotic and 2 Pseudomonas
antibiotics seem to be “too much” given the
available evidence for routine treatment of a
LTCF patient hospitalized with pneumonia

93 y.o. F from LTCF with pneumonia.
Treatment?

A.
B.
C.
D.
E.




Levofloxacin
Levofloxacin + vancomycin
Levofloxacin + Pip/Tazo + vancomycin
Ceftriaxone
Ceftriaxone + azithromycin