A 6-year-old girl developed shock and multiple organ dysfunction including acute respiratory distress syndrome in association with parvovirus B19 infection. The diagnosis was based on positive antibodies and the detection of parvovirus 19 DNA in serum, bronchial secretions and skin biopsy. It seems likely, but it was not proved, that the parvovirus infection caused acute respiratory distress syndrome.

The immune response to Mycobacterium tuberculosis is mediated by T lymphocytes. We studied the changes in lymphocyte populations occurring in peripheral blood, pleural fluid, and ascites during tuberculosis infection. For this purpose, we compared recent-onset patients (newly converted to positive Mantoux reactions) with previously diagnosed patients (individuals with organic lesions). Recent infection was associated with peripheral blood lymphocytosis involving T lymphocytes expressing either T-cell receptor ␣/␤ or ␥/␦. Lymphocytosis involved both CD4 and CD8 cells. On the other hand, we detected no changes in the distribution of peripheral blood lymphocyte populations in previously diagnosed patients. No changes were found in the numbers of B lymphocytes or natural killer cells in either recently infected or previously diagnosed patients. The pleural effusion and ascitic fluid samples contained T lymphocytes expressing T-cell receptor ␣/␤, the majority of which were CD4 ؉. These lymphocytes showed an inverted CD45RA-to-CD45RO ratio, and we found high-level expression of the interleukin-2 receptor (CD25) in some patients. The results are compatible with the existence of periods of cell activation in the pleural fluid (which are disclosed by the appearance of the CD25 antigen and the transition of CD45RA expression to CD45RO) together with nonactivation periods (loss of CD25 and persistence of CD45RO expression). We studied a fraction of the V␤ repertoire in peripheral blood in both groups and the same fraction of the V␤ repertoire in pleural fluid from patients with tuberculous pleuritis, demonstrating that, in recently infected subjects, lymphocytosis was produced by the increase in lymphocytes which expressed some specific V␤ subfamilies that differed from one individual to another. In two of five patients studied, we found significant changes in the V␤ repertoire between lymphocytes from peripheral blood and the pleural fluid samples.