The serology of pneumococcal infection: A study of the laboratory diagnosis of pneumococcal infection and the distribution of pneumococcal types

Smart, Leslie E (1987) The serology of pneumococcal infection: A study of the laboratory diagnosis of pneumococcal infection and the distribution of pneumococcal types. PhD thesis, University of Glasgow.

Full text available as:
[thumbnail of 10948145.pdf] PDF
Download (13MB)


The aim of this study was to assess the value of serology in the laboratory diagnosis of pneumococcal infection. This included the serotyping of pneumococci isolated on culture, detection of pneumococcal capsular polysaccharide antigen in body fluids and estimation of antibody responses. The co-agglutination test which uses staphylococcal protein A, was found to correlate well with the standard capsular reaction test. However, since co-agglutination was more sensitive and therefore less specific, a number of bacteria which were not pneumococci reacted in the co-agglutination test. The detection of pneumococcal capsular polysaccharide antigen in clinical material has been widely used for the diagnosis of pneumococcal infection. Antigen detection in material obtained from patients who had been treated with antibiotics provided evidence of infection. Four methods for antigen detection were compared. These were counterimmuno-electrophoresis (CIE), co-agglutination (CoA), latex agglutination (LA) and enzyme-linked immunosorbent assay (ELISA). Their sensitivity was compared in tests carried out using purified pneumococcal capsular polysaccharide and with different kinds of pneumococcal antisera. ELISA was the most sensitive method but only with antisera which contained a limited number of pneumococcal antibody specificities. Using the same antisera, CoA and LA showed similar intermediate sensitivity : CIE was the least sensitive method. However, in tests with polyvalent pneumococcal antisera CIE, CoA and LA were equally sensitive although pneumococcal antigen of types 7F, 14 and 33F could not be detected by CIE. The four methods were also evaluated to discover their suitability for the detection of pneumococcal antigen in body fluids and secretions. Co-agglutination with antiserum pools or pneumococcal typing sera was the most reliable method for the detection of pneumococcal antigen in respiratory secretions - where specificity but not sensitivity is important. Latex agglutination on the other hand was preferred for the examination of serum, urine and cerebrospinal fluid where sensitivity is the more important criterion. ELISA was a useful confirmatory test when carried out on serum or CSF which contained small amounts of pneumococcal antigen. On this basis a protocol was established and used in a comprehensive study of pneumococcal infection in hospitalised patients. Patients with respiratory and non-respiratory pneumococcal infection were classified on the basis of clinical history and the severity of disease. The distribution of pneumococci and pneumococcal antigen in clinical material obtained from different body sites was found to differ in patients in the different clinical groups. The presence of pneumococcal antigen in serum and urine was associated with severe pneumococcal disease. Since the laboratory diagnosis of pneumococcal chest infection depends, to a large extent, on examination of expectorated sputum the relationship between the results of conventional examination for pneumococci and of tests for pneumococcal antigen was studied in detail. The presence of pneumococcal antigen in sputum correlated well with clinical evidence of infection. Antigen detection proved to be a more sensitive and reliable method for the diagnosis of pneumococcal pneumonia than were the traditional techniques of culture and Gram's stain. The value of serial sputum examination in different groups of patients was demonstrated : this was able to distinguish between relapse and new infection. An indirect immunofluorescent assay (IFA) was developed for the detection of type-specific pneumococcal antibodies. Pre-absorption of the serum to remove antibody to pneumococcal C-polysaccharide improved the specificity of the assay which proved to be sufficiently sensitive to detect changes in antibody titre between acute and convalescent phases of infection. Moreover, for the majority of serotypes the minimal amount of antibody detectable by IFA was less than that currently recognised as protective against bacteraemic pneumococcal disease. Type-specific pneumococcal antibodies were often detected in acute phase sera especially in less severe infections. The presence of pre-existing type-specific antibody in serum, however, did not prevent systemic pneumococcal disease due to homologous strains of Streptococcus pneumoniae. The serotype distribution of 874 strains of S. pneumoniae was determined in relation to patients' age and to the frequency of isolation from systemic disease. Types 14 and 18 in pre-school children, and types 1,4,7,8 and 12 in patients over 5 years of age were significantly associated with systemic disease. The serotype distribution of pneumococci or pneumococcal antigen in the sputum of 1,682 patients with pneumococcal chest infection was determined. Types 3 and 8 were significantly associated with pneumonia whereas types 6 and 17 were associated with the milder clinical diagnosis of chest infection. Type 1 was common in patients with pneumonia but was rarely isolated from the sputum of other infected patients. The results of this study confirmed that serotyping of pneumococci and the detection of pneumococcal antigen in clinical material was a valuable laboratory tool for the diagnosis and assessment and also for management and prognosis of pneumococcal infection. However, interactions in vivo between pneumococcal antigen and antibody are complex. Preliminary results suggest that antibody determination may contribute to clinical assessment but further studies will be required to establish this unequivocally.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Keywords: Microbiology, Pathology
Date of Award: 1987
Depositing User: Enlighten Team
Unique ID: glathesis:1987-76652
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 19 Nov 2019 13:58
Last Modified: 19 Nov 2019 13:58

Actions (login required)

View Item View Item


Downloads per month over past year