Back to search

Democritus University of Thrace (DUTH)

Προσδιορισμός της διαγνωστικής ποιότητας και αξίας της κατά Gram χρώσης των πτυλέων σε ασθενείς με εξωνοσοκομειακή πνευμονία με τη χρήση του αλγόριθμου Gibbs: υπέρβαση της ανυπαρξίας Gold Standard

Abstract

dc:description

The aim of this study was to determine the diagnostic accuracy and diagnostic value of sputum Gram stain in etiological diagnosis and initial selection of antimicrobial therapy of bacterial community acquired pneumonia (CAP). This is a prospective study conducted through January 2002 till June 2008 in Konstantopoulio General Hospital of Athens, Greece and University Hospital of Alexandroupolis, Thrace, Greece. During this period, 1390 patients, fulfilling the diagnostic criteria of bacterial CAP were admitted, through the emergency room and the outpatient clinic. Of these 1390 patients, 178 were fulfilling the criteria for inclusion in the study. Criteria for the diagnosis of bacterial pneumonia14: • clinical findings (fever> 38.3°C and productive cough with purulent sputum production and tachypnea  25/min and/or physical findings of pulmonary consolidation on examination) • WBC 15.000/μL with polymorphonuclears 75% • chest x-ray showing homogeneous or inhomogeneous infiltrates, with or without segmental distribution and/or pleural effusion • in every patient, on admission and before starting antimicrobial therapy, were taken 6 blood cultures (3 aerobic and 3 anaerobic) from three different peripheral veins and sputum for culture and Gram stain. Criteria for inclusion in the study: 1) no antimicrobial therapy within the past two weeks, 2) no clinicolaboratory evidence of infection in another site: urinary tract infection (negative urine culture), sinusitis (negative X-ray), central nervous infection (negative spinal tap whenever it was indicated), endocarditis (echo negative for vegetations), arthritis, pharyngitis, osteomyelitis) and 3) isolation of the same microorganism from the blood and sputum cultures (gold standard). Sputum specimens were collected upon patient’s admission by an experienced resident and send to the laboratory immediately. Gram stain was also done and read upon admission, by two experts, who were blind, as far as, blood cultures and sputum cultures results were concerned (which at any rate were available after two to four days). Criteria of validity of sputum specimens: Epithelial cells/ low field power SCORE Polymorphonuclears /low field power SCORE >25 -3 >150 +3 16-25 -2 76-150 +2 5-15 -1 1-75 +1 Criteria of positivity of Gram stain: >50% microorganisms of same morphotype. Given the above mentioned criterion, the presence of: Gram positive diplococci (Gr+DC) was suggestive for Pneumococcal pneumonia; Gram positive cocci in clusters (Gr+CC) suggestive for Staphylococcal pneumonia; Gram negative coccobacilli (Gr–CB) suggestive for H. influenza pneumonia; Gram negative bacilli (Gr–B) suggestive for Gram– bacilli pneumonia. Taking as gold standard the blood and sputum cultures results, we calculated: 1) the prevalence (P: frequency of the target disease in the population of interest) of each etiological diagnosis of CAP, and 2) the diagnostic value (point estimate and 95% confidence interval), 4) the positive and negative diagnostic gain (G+, G–) and expected diagnostic gain (G), and 5) the concentration and reduction factor (CF, RF), of the sputum Gram stain for the etiological diagnosis of CAP. For the calculation of the diagnostic quality and diagnostic value of the sputum Gram stain, in the absence of gold standard, one of the Markov Chain Monte Carlo simulation techniques, the Gibbs sampler algorithm, was used. For the implementation of this method was used a computer program written in S-PLUS. The following results, in the form of medians and 95% confidence intervals were calculated: The 95% confidence interval of the prevalence of pneumonia was 0.50 to 0.63 (median 0.56) for the streptococcal pneumonia, 0.10 to 0.18 (median 0.13) for the staphylococcal pneumonia, 0.12 to 0.21 (median 0.16) for the pneumonia from H. Influenzae and 0.12 to 0.22 (median 0.17) for the pneumonia from Gram negative bacilli. The 95% confidence interval of the sensitivity of sputum Gram stain was 0.57 to 0.72 (median 0.65) for the streptococcal pneumonia, 0.58 to 0.91 (median 0.77) for the staphylococcal pneumonia, 0.69 to 0.93 (median 0.83) for the pneumonia from H. Influenzae and 0.53 to 0.89 (median 0.73) for the pneumonia from Gram negative bacilli. The 95% confidence interval of the specificity of sputum Gram stain was 0.89 to 0.98 (median 0.94) for the streptococcal pneumonia, 0.91 to 0.98 (median 0.95) for the staphylococcal pneumonia, 0.95 to 0.99 (median 0.98) for the pneumonia from H. Influenzae and 0.87 to 0.93 (median 0.90) for the pneumonia from Gram negative bacilli. The 95% confidence interval of the positive likelihood ratio of sputum Gram stain was 5.50 to 28.70 (median 11.22) for the streptococcal pneumonia, 7.65 to 39.57 (median 15.74) for the staphylococcal pneumonia, 17.32 to 139.68 (median 41.49) for the pneumonia from H. Influenzae and 4.87 to 11.16 (median 7.54) for the pneumonia from Gram negative bacilli. The 95% confidence interval of the negative likelihood ratio of sputum Gram stain was 0.29 to 0.46 (median 0.37) for the streptococcal pneumonia, 0.09 to 0.44 (median 0.24) for the staphylococcal pneumonia, 0.07 to 0.32 (median 0.17) for the pneumonia from H. Influenzae and 0.12 to 0.52 (median 0.30) for the pneumonia from Gram negative bacilli. The 95% confidence interval of the positive predictive value of sputum Gram stain was 0.85 to 0.97 (median 0.92) for the streptococcal pneumonia, 0.53 to 0.89 (median 0.73) for the staphylococcal pneumonia, 0.73 to 0.96 (median 0.88) for the pneumonia from H. Influenzae and 0.27 to 0.52 (median 0.39) for the pneumonia from Gram negative bacilli. The 95% confidence interval of the negative predictive value of sputum Gram stain was 0.63 to 0.78 (median 0.71) for the streptococcal pneumonia, 0.91 to 0.99 (median 0.96) for the staphylococcal pneumonia, 0.94 to 0.99 (median 0.97) for the pneumonia from H. Influenzae and 0.95 to 0.99 (median 0.97) for the pneumonia from Gram negative bacilli.

Degree

thesis:*
Grantor dc:publisher
Democritus University of Thrace (DUTH)
Year dc:date
2009

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Anevlavis, Stavros
  • Ανευλαβής, Σταύρος

Subjects

dc:subject × 14

Rights

Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/28127

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Anevlavis, Stavros; Ανευλαβής, Σταύρος. Προσδιορισμός της διαγνωστικής ποιότητας και αξίας της κατά Gram χρώσης των πτυλέων σε ασθενείς με εξωνοσοκομειακή πνευμονία με τη χρήση του αλγόριθμου Gibbs: υπέρβαση της ανυπαρξίας Gold Standard. Democritus University of Thrace (DUTH), 2009. http://hdl.handle.net/10442/hedi/28127