Observations on Pyuria in Children

The diagnosis of renal tract infection rests upon the recognition of symptoms and the demonstration of pyuria. The bacteriology of catheter specimens of urine can support this diagnosis but can never form its basis as contamination occurs so readily. In acute infections symptoms are usually prominent, but in chronic pyelonephritis, where the infection may be low grade but persistent, symptoms may be miinimal. The diagnosis in such cases depends on the demonstration of pyuria which may be modest and intermittent. To know the difference between a high normal excretion of leucocytes in the urine and a low-grade but significant pyuria is therefore crucial. It is our belief that the customary method of measuring pyuria by counts per high or low power microscope field is so crude that only gross amounts can be distinguished, and an undue reliance on the method will lead to many cases of renal tract infection being overlooked. There is, moreover, a lack of agreement in interpreting results when only a few cells are present. Campbell (1951) writes in his textbook that 'there is no accepted standard of what constitutes an abnormal amount of pus in the urine specimen as commonly examined. Considerable difficulty may arise in differentiating between a high normal exudate of leucocytes and mild pyuria.' He considers that in an uncentrifuged catheterized specimen 3 to 5 cells per low power field are within normal limits. Craig (1935) could not find more than I cell in 10 to 15 high power fields when examining urines of newborn infants. On the other hand, many authors believe that at least 6 to 8 cells per high power field should be present before the diagnosis of significant pyuria is justified. Much of the literature on the subject has been reviewed by Hepler and Scott (1935) who point out that various arbitrary limits of normal are given with no indication of the method of collection or examination, and that there is no report of a systematic study that would definitely confirm or refute many of the assumptions that appear in the literature on the subject.