Burden of Hospitalized Pneumonia in Korea COPD Population (COPDCAP)

SUBJECTS AND CENTERS A. Subjects; COPD patients hospitalized for community-acquired pneumonia from Feb, 2017 to Feb, 2018.

B. Centers; More than five university-affiliated hospitals of Korea

STUDY DESIGN

; prospective, multi-center, cohort study

STUDY ANALYSIS A. The required number of COPD patients hospitalized for community-acquired pneumonia

• The sample size calculations were based on detecting 1 point difference in CURB-65 score between influenza/pneumococcal vaccine recipients and non-recipients (expecting the mean CURB-65 score of 3 points for influenza/pneumococcal vaccine recipients and 4 points for non-recipients), assuming a common standard deviation of 1 point in each group,* a two-tailed test at 5% type I error, and a desired power of 80%. In terms of PCV 13, the estimated sample size for an independent t-test is 90 when we expect that 10% of them were PCV13 recipients and the other 90% were non-recipients (in Korea, 10% of candidates are estimated to be vaccinated with PCV13). Since the CURB-65 scores may not be symmetrically distributed, adding 15% to the required number was considered for a non-parametric test and the desired number of patients with COPD and pneumococcal pneumonia is 104 for the analysis. Total 384 COPD patients admitted for community-acquired pneumonia will be enrolled, assuming 30% prevalence of pneumococcal pneumonia among them (the prevalence of pneumococcal pneumonia is 25~45% in Korea), with allowance of 10% possible data loss.

  • referenced from 'Role of Semi-quantitative Serum Procalcitonin in Assessing Prognosis of Community Acquired Bacterial Pneumonia Compared to PORT PSI, CURB-65 and CRB-65 Journal of Clinical and Diagnostic Research. 2015 Jul, Vol-9(7): OC01-OC04' B. Parameters to be evaluated A) Demographic data . Age, gender, BMI, smoking status, socioeconomic status B) Past medical history

    . Comorbidities; diabetes, hypertension, malignancy, chronic renal disease, chronic liver disease…

    . Smoking status and amount of smoking by pack-year

    . Vaccination status (types and timing)

    - Will be checked both by history taking and by the review of medical record

    - PCV13, PPV23, Influenza vaccine, Zostavax…. C) Diagnostic tests for the identification of pneumonia pathogens

    . Sputum Gram stain/culture (adequacy will be evaluated with the Bartlett criteria)

    • Sputum PCR test for bacterial pneumonia pathogens

      • Multiplex PCR test designed for the detection of S. pneumoniae, M. pneumoniae, L. pneumophila, H. influenza, B. pertussis, C. pneumoniae

    • Urine antigen test(Binax and SS-UAD) for pneumococcus - two aliquots of the urine sample will be frozen and stored for subsequent shipment to the central laboratory (Pfizer Vaccines Research East and Early Development, Pearl River, PA, USA) for Luminex platform-based multiplex UAD assay testing as described previously (Pride et al. Clin Vaccine Immunol 2012;19:1131-41). Processing, storing and shipping of urine samples will be carried out according to Pfizer protocol. Urine will be tested locally using the commercially-available BinaxNOW® S. pneumoniae antigen test also.

      • Control urine samples will be collected from 400 adults without pneumonia according to the Pfizer urine collecting method and stored for subsequent shipment to the central laboratory (Pfizer Vaccines Research East and Early Development, Pearl River, PA, USA) for Luminex platform-based multiplex UAD assay testing.
    • Two sets of blood culture D) Severity scores
    • Pneumonia severity index (PSI)
    • CURB-65 E) Outcome parameters
    • ICU admission rate
    • Mechanical ventilation rate
    • ICU stay and hospital stay
    • Mortality

SAFETY;

; This is an observational study which evaluates the effects of previously inoculated influenza/pneumococcal vaccine. Treatment for pneumonia and COPD will be done by attending physician with their own disposal. In this setting there would not be ethical issues to be debated.

DECISION POINTS/ STATISTICAL METHODS/INTERIM ANALYSIS

; The statistics will be purely descriptive. Continuous variables will be presented as mean ± SD or median value with minimum and maximum values. Categorical data will be presented as frequency and percentile. Intergroup comparison of continuous variables will be done with student T test and that of categorical data will be done with Chi-square test or with Fisher's exact test. Statistical significance will be determined at the 5% level. We will consider non-parametric equivalent tests if the distributional assumptions are not suitably made.