EBM

May 29, 2009

Grant Paulsen

Clinical Question: Does every patient admitted from a nursing home need Nosocomial PNA treatment?

Effect of Antibiotic Guidelines on Outcomes of Hospitalized Patients with Nursing Home-Acquired Pneumonia. J Am Geriatr Soc. 2009 Apr 30.

Study Design: Retrospective.

Background:Pneumonia is the leading infectious cause of death in patients 65 and older. The yield of diagnostic tests is low (<50%), therefore most treatment is empiric and directed by guidelines. The two most recent guidelines are similar, but with slightly different focuses. The 2003 IDSA guidelines for treatment of PNA places less emphasis on drug-resistant organisms than the 2005 American Thoracic Society/IDSA guidelines. This article sought to evaluate the effect of guideline based empiric therapy for nursing home acquired PNA on time to clinical stability, time to switch therapy, length of stay and mortality.

Study Population: 334 consecutive adult patients admitted to the study hospitals (NY state) with principle diagnosis of pneumonia and residing in a nursing home, between Jan 2006 and Aug 2008.

Inclusion Criteria: As above, with diagnosis of pneumonia requiring that the patients have at least two of the following:

-new or increased cough,

-new or increased sputum production

-temp of 38.0 or higher, or 35.0 or lower

-pleuritic chest pain

-new or increased shortness of breath

Plus

-new or worsening infiltrates on CXR within 48hrs of presentation

Characteristics on Table 1.

Exclusion Criteria: Patients were excluded if:

-Received more than one dose of antibiotics prior to presentation

-HIV, or AIDS

-Hospitalized in last 30 days

-Admitted to the ICU (from the ED)

-OSH transfers

-Active chemotherapy

Outcomes:

-Clinical Stability, defined as meeting all parameters of:

  • T 37.2 or less
  • HR <100
  • RR <24
  • SBP >90
  • PaO2 60mmHg off supplemental O2

-Time to switch from IV to PO therapy

-Mortality

Statistical Analysis:Intention-to-treat analysis, accomplished by considering death as failure to meet criteria for clinical stability, switch therapy, or discharge and those patients were arbitrarily assigned 7 days as time to stability, and switch therapy. Independent variables were assessed using multivariate linear regression.

-To ‘control’ for selection bias, the authors constructed a propensity score.

  • ‘In statistics, propensity score matching (PSM) is one of quasi-empirical “correction strategies” that corrects for the selection biases in making estimates.’ Wikipedia.

Results:

-334 patients, 258 (77%) were treated according to 2003 CAP Guidelines, and 76 (23%) were treated according to 2005 HCAP guidelines.

-71 were excluded for receiving antibiotics that didn’t conform to either guideline

-Baseline characteristics showed no statistical difference in both groups, Table 1.

-Quality indicators for NHAP were similar in both groups, Table 2.

-Treatment regimens:

  • 2003 CAP – -lactam (CTX or Amp/Sulbactam) + macrolide [29% of patients], or fluoroquinolone monotherapy (moxifloxacin) [71%].
  • 2005 HCAP – Vanc, antipseudomonal PCN and antipseudomonal fluoroquinolone [89%].
  • 11% were given a monobactam due to PCN allergy.

-Time to clinical stability was not significantly different between both groups; 3.6 +/- 0.7 days in 2003 CAP group, vs. 3.9 +/- 0.8 days in 2005 HCAP (P = 0.38).

-Also no difference in ICU admission, development of septic shock, or mortality:

  • In-Hospital mortality – 13.2% in 2003 CAP group vs. 17.1% in 2005 HCAP, P = 0.49. OR 0.73, CI = 0.36-1.48 (95% CI)
  • 30-Day mortality – 15.1% in 2003 CAP group vs. 22.4% in 2005 HCAP group, P = 0.19. OR 0.66, CI = 0.35-1.25.

-Time to switch to oral therapy: 4.1 +/- 1.6 days in 2003 CAP vs. 5.8 +/- 2.2 days in 2005 HCAP group, P <.001

-Length of hospital stay: 5.7 +/- 1.7 days in 2003 CAP vs. 6.9 +/- 2.8 days in 2005 HCAP group, P<.001.

-No significant differences occurred when those patients that died were excluded, or when the propensity score was included.

-Multivariate analysis demonstrated that the factors associated with delay in achieving clinical stability were Pneumonia Severity Index (P=.006), and multilobar involvement (P=.005).

  • PSI assigns a severity class (I-V) based on demographics, comorbidities, physical exam, and labs
  • Basically word-for-word PORT Scoring

Validity:

-Were the patients randomized? No, retrospective study

-Were patients analyzed in the groups in which they started? Yes

-Were the groups similar at the start of the study? No

  • No ‘statistical difference’ of p<0.05, but overall both groups were not the same

-Does the propensity score & multivariate analysis remove selection bias? No

-Is the study powered enough to answer the question? No

-Can I use this study to make a clinical decision? No

Strengths:

-Multi-centered (albeit all in the same state)

-Intention-to-treat analysis

Weakness:

-Under-powered

-Retrospective, chart-review study; not randomized.

-Cannot effectively control for selection bias.

-Regional study only

Discussion:

** How were the 2005 ATS/IDSA pneumonia guidelines developed?

- The ATS/IDSA Guideline Committee originally met as a group,with each individual being assigned a topic for review and presentationto the entire group. Each topic in the guideline was reviewedby more than one committee member, and after presentation ofinformation, the committee discussed the data and formulatedrecommendations. Two committee members prepared each sectionof the document, and a draft document incorporating all sectionswas written and distributed to the committee for review andsuggestions. The guideline was then revised and circulated tothe committee for final comment. This final statement representsthe results of this process and the opinions of the majorityof committee members.

-Relied heavily on microbiology data, with very little outcome data available.

  • Large percentage from VAP studies

-Algorithms were then developed. See Below.

**What is the evidence of the effectiveness of these guidelines? None that I could find

2005 ATS/IDSA pneumonia treatment algorithm:

TABLE 2. Risk factors for multidrug-resistant pathogens causing hospital-acquired pneumonia, healthcare-associated pneumonia, and ventilator-associated pneumonia

• Antimicrobial therapy in preceding 90 d
• Current hospitalization of 5 d or more
• High frequency of antibiotic resistance in the community or
in the specific hospital unit
• Presence of risk factors for HCAP:
Hospitalization for 2 d or more in the preceding 90 d
Residence in a nursing home or extended care facility
Home infusion therapy (including antibiotics)
Chronic dialysis within 30 d
Home wound care
Family member with multidrug-resistant pathogen
• Immunosuppressive disease and/or therapy