|
|
Benefits of quantitative sputum processing in diagnosis of respiratory tract infections
KOUBOVÁ, Lucie
Lower tract respiratory infections (LRTI) remains worldwide among one of the most common causes of morbidity and therefore the early diagnosis and timely onset of effective antibiotic therapy is very important. The most often non-invasive material from LRT sent for microbiological analysis in suspected infections of LRT is spontaneously expectorated sputum. Basic method for its processing is then microscopy and cultivation analysis. For laboratory and consequently for the attending doctor to obtain the most exact correlation between the laboratory interpretation of the result and the clinical status of the patient is to asure clinical as well as technical validity of the sample. The purpose of this work is the comparison of two bacteriological methods for processing the sputum samples ? classic cultivation method of undiluted sputum samples and quantitative sputum culture. Another aim is to evaluate whether the quantitative sputum culture when only purulent samples are cultivated allows microbiologist higher quality interpretation of the analysis result. Clinical samples of spontaneously expectorated sputum were processed at the Department of Clinical Microbiology and Antibiotic Station (KMAS), Central Laboratories, Strakonice Hospital from April to October 2012. During the six months period KMAS received in total 266 samples of sputum, out of which 230 came from the hospitalized patients and 36 came from primary and outpatients. Sputum samples were received from the patients suspected of LTRI and in most of the cases sputum was collected prior to antibiotic therapy.All the samples were analysed by both compared methods. In classic cultivation method simultaneously with the microscopic Gram analysis the sample was inoculated on selected culture media. During the quantitative sputum culture the microscopic Gram analysis was performed first and contaminated sputa coming from URT were excluded from further analysis. Quantitative sputum culture was then performed only on purulent sputum samples with low rate of squamous epitelial cells. Those were homogenised prior the cultivation itself. The results of both methods were compared with the clinical status of the patient and then clinical characteristics of the methods were statistically analysed. They allowed the assessement of the precision of a particular method. All 36 samples from outpatients and primary care patients were processedby classic cultivation method. Sensitivity of classic cultivaiton method versus quantitative culture method reached 100%/ 100%, respectively, specificity reached 69%/74%, efectivity 69%/75%, positive predictive value 3%/3%, negative predictive value 100%/100% and positive credibility of laboratory analysis equaled 3,18/ 3,89. Out of the total 266 recieved samples from hospitalized patients 130 samples (57%) were processed by quantitative sputum culture on the basis of microscopic Gram analysis. Sensitivity of classic cultivation method versus quantitative culture method for these samples reached 62%/ 72%,, respectively, specificity reached 54%/67%, efectivity 56%/68%, positive predictive value 12%/14%, negative predictive value 54%/67% and positive credibility of laboratory analysis equaled 1,13/ 2,16. On the basis of the obtained results it is possible to recommend the quantitative sputum culture as the method which improves the interpretaion in the diagnostic of LRTI in all hospitalized, outpatient and primary patients. By excluding the material which is not suitable for quantitative sputum culture is not only shortened the time for the result interpretation but also consumer material is safed in the laboratory. The sputum samples cannot be excluded exclusively on the basis of microscopy analysis. Diagnosis of the patient must to be taken into account as well. Quantitaive sputum culture in microscopically valid samples allows microbiologist predict with greater accuracy colonising suspect pathogens and isolates truly pathogenic.
|
guest ::
