Bacterial pneumonias--evaluation of various sputum culture methods.MP Verenkar, MJ Pinto, R Savio, N Virginkar, I Singh
Dept of Microbiology, Goa Medical College, Bambolin.
With an objective of improving diagnostic value of sputum in bacterial pneumonias, 50 uncomplicated 'community' acquired cases were studied using Gram staining of sputum along with bedside inoculation with/without dilution of the specimen. Gram staining of sputum samples collected before treatment revealed pneumococcal infection in 46% cases. The results were however inconclusive on samples sent by routine procedure involving logistic delay. Cultural analysis of sputum processed by three different techniques showed that bedside inoculation of sputum after dilution to be the most efficient technique yielding Streptococcus pneumoniae in 34% cases, Gram positive cocci in lesser number (20%), Gram negative rods (GNR) in 18% cases. Sputum samples processed bedside without dilution yielded a lower number of pneumococci and other Gram positive cocci (24% & 16% cases respectively). Routine processing of sputum, involving logistic delay yielded a high number of Gram negative rods (62%), indicating their overgrowth. Thus bedside inoculation of sputum after dilution coupled with direct Gram staining serves as a simple and yet valuable laboratory aid in the diagnosis of uncomplicated 'community' acquired bacterial pneumonias.
Keywords: Bacterial Infections, diagnosis,microbiology,Bacteriological Techniques, Gram-Negative Bacteria, isolation &purification,Gram-Positive Bacteria, isolation &purification,Human, Pneumonia, diagnosis,microbiology,Sputum, microbiology,
Despite the great development, which has taken place in the chemotherapeutic fields, pneunonia still remains an infectious disease with high mortality and morbidity. Correct diagnosis of the etiological pathogen depends on correlation of many factors including patient history, clinical presentation and laboratory data. The most common laboratory aides in persuit of the causative agent include microscopy and culture of sputum in addition to blood culture. However, interpretation of these data can be problematic because of the ubiquitous mixed flora and frequency of carrier rate for the respiratory pathogens, in addition to the difficulties in obtaining appropriate specimens, which are representative of bronchial secretions rather than saliva.
To overcome these problems and to improve the diagnostic value of sputum, various means such as selection of purulent material from sputum either directly or by washing, by homogeniation or by quantitative cultures have been attempted Recently, a number of invasive procedures like transtracheal aspiration, brush biopsy and lung aspirate have been recommended which prevent contamination of specimen with flora of upper airways. However, these techniques have their own limitations requiring sophisticated equipments and expertise.
Coupled with these inherent problems, in our set-up in Goa, is the logistic factor introduced since 1982. The Microbiology laboratory was moved 7 kms away from the main Hospital in Panaji, as the initial phase of shifting the whole Medical College Complex to Bambolin.
Thus, with the objective of improving the diagnostic value of sputum in bacterial pneumonias, uncomplicated community acquired cases were studied making use of Gram staining of sputum and bedside inoculation with and without dilution of the sputum sample to identify the pathogen.
This study includes 50 cases of uncomplicated community acquired pneumonias admitted to the Tuberculosis and Chest Diseases Hospital, Goa Medical College during one year period (March 1987 February 1988). Following a complete clinical and radiological examination, sputum and blood samples were collected before administering antibiotics.
The sputum was collected in sterile ice - cream cups after instructing the patient to rinse his/her mouth thoroughly with water and cough forcefully to bring out the material from the tracheobronchial tree.
Gram staining was done in the side laboratory, where material, media and equipment were kept ready. Purulent portion of the sputum was used for smear preparation and graded as per Murray and Washington grading system for assessing the quality of the samples as follows:
Only grade 5 samples were selected for further study as the large number of epithelial cells in grades 1 to 4 indicate qontamination with oropharyngeal secretions and invalidates the sample. In such a case, the patient was asked to produce another sample, which was then subjected to similar study.
After Gram staining the purulent portion was cultured immediately on blood agar and MacConkey plates. An optochin disc was used if the staining indicated presence of pneumococci.
The remaining sample was then subjected to liquefaction and dilution as described by Dixon and Miller using N-acetyl-I-cysteine. Cultures of these diluted samples were made on blood agar, chocolate agar and MacConkey agar. The plate were incubated at 37?C in the side laboratory incubator in the atmosphere of 5 10% CO2. After overnight incubation the plates were transferred immediately to the Microbiology department where predominant organisms were studied by standard techniquese. Optochin sensitive organisms were studied for further confirmation of Streptococcus pneumoniae. All pathogenic organisms were subjected to antibiotic sensitivity testing by Kirby Bauer technique.
Simultaneously, before treatment, the concerned Department of Tuberculosis and Chest Diseases sent a sputum sample to the Microbiology Department for bacteriological studies as was being done routinely. Following Gram staining of the samples, cultures were made on blood agar and MacConkey agar and read after overnight incubation at 37?C.
Besides sputum samples, blood cultures were made in glucose broth and taurocholate broth at the time of admission and processed according to standard techniques. However, the data of blood cultures is not analysed in this study.
The sputum samples collected before treatment on Gram staining revealed pneumococci in 23 patients
(46%) while inconclusive in 13 cases (26%) where mixed flora was seen (Table-2). Gram staining results of most of the sputum samples sent by the concern departments on routine bases, showed Gram negative rods (GNR), Gram positive cocci and rods in large numbers.
Cultural analysis of sputum samples processed differently with application of three techniques showed significantly different results. Modified method of Dixon and Miller using N-acetyl-1 cysteine with bedside inoculation was found to be the most efficient method for isolatation of pathogens from sputum. Isolation rates of various organisms using all three methods are illustrated in Table-3.
Streptococcus pneumoniae d be isolated from 17 cases (34%) in whom, the sputum was subjected to liquefaction before bedside inoculation, from 12 cases (24%) in whom sputum samples were processed at bedside without liquefaction and only from 3 samples that were transported to the laboratory. Rate of isolation of other Gram positive cocci was also higher with samples processed with or without liquefaction.
Isolation of GNRS was however, higher in samples directly sent to the laboratory for processing (31 patients; 62% and included Klebsiella species),
Escherichia More Details coli and Pseudomonas species. Samples inoculated at bedside before and after liquefaction showed a lower isolation rate of GNRS (28% and 18% respectively).
Normal flora was grown in 14 sputum samples (28%) processed at bedside after liquefaction and in 16 (32%) without liquefaction. Samples sent to the laboratory yielded normal flora in only 10 cases (20%).
Pneumonia is a common and important respiratory infection in the community, pneumococcus being the commonest etiological pathogen.
Microscopic examination is useful in presumptively identifying the etiological pathogen especially in pneumococcal pneumonias. In our study, Gram staining was suggestive of pneumococcal etiology in 23 patients(46%). Rein et al have shown sensitivity and specificity of Gram staining to range widely with overall accuracy of approximately 62% in identification of pneumococci in sputum.
In addition, microscopic examination of sputum serves as an important guide to assess the suitability of the sample for cultural studies. This is based on cellular composition of the sputum. Presence of less than 10 epithelial cells and more than 25 leucocytes per low power field in the samples makes it a true representative of bronchial secretions, and thereby improves the diagnostic value of culture. Thorsteinsson et al have shown that a properly collected sputum based on these guidelines is as accurate as a transtracheal or bronchial aspirate in the diagnosis of acute pneumoccal pneumonias.
However, interpretation of Gram staining presents certain problems because of the presence of ubiquitous flora in the upper respiratory tract. In addition, Gram negative organisms are often overlooked and considered as non-bacterial elements, which also stain. Gram negative while Gram positive debris may be mistaken for organisms.
As sputum is often contaminated with organisms from oropharynx, in order to facilitate isolation identification of bacterial agent, Dixon and Miller recommended a dilution technique based on the principle that true pathogens always exist in a significant count in the sputum and will not be diluted out completely by dilution, while the commensals would be diluted and hence reduced to a minimum. This was proved beyond doubt in our study as the isolation rate of pathogens from diluted samples was distinctly higher. Moreover, quefaction brings about homogenization of samples and avoids the risk of accidental choosing of a nonpurulent portion of the specimen. The only drawback of this technique is that negative results may be obtained in partially treated cases due to presence of pathogens in few numbers in the sputum samples.
Another factor that vitiates the culture results is the delay in processing of the samples after its collection. Undesirable but usual delay in transportation of material to the laboratory permits overgrowth of commensal bacteria and results in failure to isolate the etiological agent. This clearly explains the high isolation rate(62%) of Klebsiella species and other GNIRS from sputum samples processed in our laboratory.
Considering the fact that the cases under study were uncomplicated community acquired pneumonias, one would incriminate Gram positive bacteria as the etiological pathogens. Hence the isolation of GNRS from sputum samples cultured in the laboratory after logistic delay indicates their overgrowth. Bedside inoculation or inoculation with least delay is hence recommended to overcome these problems.
In our study bedside inoculation of sputum after liquefaction appears to be the more suitable technique and when correlated with a direct Gram staining served as valuable laboratory aide. We recommend its use in centres where facilities for sophisticated instrumentation and expertise are not available.
[Table - 1], [Table - 2], [Table - 3]