Bacteria and Serologic Reactions against Bacteria in Patients Hospitalized with Acute Respiratory Illness

Abstract

During 1963–1967 a study was undertaken of 1128 patients hospitalized with acute respiratory illness (cases of scarlet fever excluded). Pneumonia was diagnosed in 53%. Bacterial cultures were made on each of the first 3 days from nose, nasopharynx, and throat swabs, and sputum. Tests for antistreptolysin, antistaphylolysin, antipneumolysin, and complement fixing antibody against H. influenzae were performed on acute and convalescent serum samples. A number of complement fixation tests against respiratory viruses, Mycoplasma pneumoniae, and Bedsonia agent were also carried out. By repetition of cultures it was found that: 1) single throat swabs gave positive results in less than half the patients harbouring streptococci; 2) nasal swabs were the best material for finding S. aureus but single samples even from all the localities mentioned above revealed less than half the patients from whom staphylococci were isolated; 3) sputum was the best material for recovering pneumococci but, in spite of the 3 cultures generally made, did not reveal more than half the patients found to harbour pneumococci; 4) sputum was also the best material for isolation of H. influenzae, but did not reveal more than about half the carriers. The coincidence between bacterial findings and the corresponding serologic reaction was far from complete. In all, potentially pathogenic bacteria were found in 43% of pneumonia patients but only in 35% of those without pneumonia and, further, in 48% of patients not treated with antibiotics prior to admission, as against 32% of those so treated. Serologic indication of bacterial infection was found in 45% of pneumonia patients as against in 28% of non-pneumonia patients but no difference was noted between patients not treated with antibiotics prior to admission and those so treated. Multiple antibody reactions against more than one bacterial species or against bacteria together with virus, Mycoplasma pneumoniae, or Bedsonia antigens did occur but, with a few exceptions, which are discussed, not more often than could be expected by random incidence. Reasons are given for a preference of serologic reactions before bacterial findings. Some epidemiologic and clinical data are discussed in relation to the bacterial findings.