Myalgic encephalomyelitis.

Sir, Apparently 'sterile' pyuria defined as the presence of more than 10 white cells MI-' of uncentrifuged urine with no growth on a primary isolation medium after overnight aerobic incubation is not an uncommon finding in children. Little attention has been paid to its significance and many laboratories no longer undertake urine microscopy, regarding culture as the only relevant test for infection. In the past, pyuria has sometimes been dismissed as a nonspecific response to fevers in children. Were this so it should occur as often in boys as in girls. In this laboratory sterile pyuria was found in 194 girls and seven boys in 1988 and 290 girls and 19 boys in 1989. This suggests that it is, in fact, indicative of an inflammatory response to infection. There are three possible explanations for this. The child may have been receiving antibacterial therapy when the specimen was collected, there may be inflammatory changes in the bladder or kidneys persisting after antibacterial treatment of an infection or there may be infection with an organism which is not detected by overnight incubation on a primary isolation medium. The latter could involve the urinary tract (the only child who has developed renal failure owing to reflux nephropathy in this district during the past 12 years was infected with Haemophilus influenzae) or, possibly, the vulva or vagina. In none of these circumstances should the pyuria be ignored. Our practice is to request a further carefully collected fresh specimen from such children, asking for information about treatment. These specimens are cultured by methods capable of detecting fastidious organisms.' Some are completely negative on microscopy 'and culture, some yield aerobic pathogens, while others yield fastidious organisms, tne nature ot wnicn suggests tnat a wiaer interest in this problem is warranted. Between August 1988 and January 1990 inclusive we have isolated fastidious organisms from the urine of 80 children, all girls (age range two to 12 years), who presented with urinary symptoms and in whom the initial specimen showed 'sterile' pyuria. The commonest isolate was Gardnerella vaginalis (39); there were 27 isolates of various fastidious species of streptococci, three of Corynebacterium sp, two ofH influenzae and nine isolates which did not identify on the basis of the tests used. It is recognized that fastidious streptococci may be pathogenic in damaged renal or bladder tissue.2'3 The predominance of G vaginalis in this series, however, is striking. This organism is increasingly recognized as a urinary pathogen in adults. It has been demonstrated in renal and bladder urine from patients with urinary tract pathology3 and in bladder wall biopsies from patients with the inflammatory changes of so-called 'interstitial cystitis'.4 It is known to be present in the vaginal flora of women and sometimes in the urethral flora of adults of both sexes. Its presence, however, was not recorded in studies of the periurethral or urethral flora of pre-menarchal girls,5'6 although in the latter study it is possible that the Corynebacterium sp recorded may have included it. Bartley and colleagues7 found it in the vaginal flora of 20 girls who had been sexually abused, although they also found it in five girls in a control group of similar size. Bacteraemia with this organism has been reported in a four week old baby girl with a congenital renal tract abnormality (PHLS, Communicable Disease Surveillance Centre, unpublished report); infection in this case was presumably via the genital tract of the mother during childbirth. The risk of urinary tract infection in children as a consequence of sexual abuse is self evident, and it is surprising that there is little published evidence on the suoject. wnen tne patnogen is a common one, such as Escherichia coli, there is no easy way to distinguish the small number of children who have been abused from the majority who suffer urinary tract infection as a result of the more generally recognized causes. However, the presence of a pathogen not usually associated with urinary tract infection in children, and not known to be present in their commensal flora, might be an important pointer to sexual abuse. Goldenring and colleagues8 reported urinary tract infection owing to Staphylococcus saprophyticus, another pathogen usually associated with urinary tract infection in young sexually active women, in a sexually abused girl of 29 months. In three of the patients reported here from whom G vaginalis was isolated the general practitioner volunteered a history of known sexual abuse. To draw this conclusion in all cases would be quite unjustified but the problem certainly warrants further investigation. General practitioners should be alerted to the importance of 'sterile' pyuria and initiate appropriate investigations and referral.