Objective To determine the effect of the International Society of Paed
iatric Oncology (SIOP) Wilms' tumour protocols (pre-operative chemothe
rapy based on clinical and radiological findings, with no prior tumour
verification by open or needle biopsy) on subsequent intra-operative
tumour diagnosis and staging. Patients and methods The diagnosis and s
taging of possible Wilms' tumour by clinical, ultrasonographic and rad
iological assessment were compared with the intra-operative findings i
n 14 consecutive children (1-12 years of age) treated between 1989 and
1994. Results A diagnosis of Wilms' tumour was histologically verifie
d in 11 of 14 children. In two children, verification was not possible
due to complete necrosis of the tumour following pre-operative chemot
herapy. In a 12-year-old boy with an estimated stage IV disease due to
a solitary lung metastasis, a renal cell carcinoma was revealed in th
e nephrectomy specimen while subsequent thoracotomy revealed dysmorphi
c but not malignant tissue, The estimated tumour stage was correct wit
h regard to localized or metastatic disease in nine of 11 children wit
h histologically confirmed Wilms' tumour, while in two children with a
n estimated stage II tumour, liver metastases were found intra-operati
vely and the tumour was upstaged to IV. Conclusion Exact tumour diagno
sis and staging was difficult in these patients, Although the accuracy
of tumour staging depends on the sensitivity of radiological and ultr
asonographic examinations, difficulties in tumour diagnosis may be ove
rcome by biopsies of the primary tumour. The justification of upstagin
g a low-stage I/II tumour to stage III, provoking a more intense post-
operative treatment, should be proven by prospective randomized studie
s. The decision to perform a primary tumour biopsy would be facilitate
d if possible subsequent deterioration of outcome could be excluded.