Background and aims: Treatment of Nephroblastoma (Wilms' tumor) in children is one of medicine's success stories. Due to improvements in surgical techniques, drug therapies, and radiation, 85 to 90 percent of children with Wilms tumor who receive state-of-the-art treatment are cured. While this is true for developed countries the outcomes of nephroblastoma (Wilms' tumor, WT) in our hospital and other developing countries are notably below those in these countries. There are continuous reviews and updates of clinical pathologic features, treatment designs, and treatment outcomes every 2 or 3 years, in the developed countries. However, in resource poor countries, such frequent reviews and updates are presently not affordable but, nevertheless, efforts should be made to undertake them every 5 or so years, to improve patient management and treatment outcomes. The objectives of this study were to review the clinical presentation and management of children with Wilms tumor and the factors influencing the outcome at Kenyatta National Referral and Teaching Hospital (KNH). This was a retrospective descriptive cross-sectional study. Analysis of the data was carried out using the statistical package for the social sciences (SPSS) software. The findings of the study will form a basis for revision of treatment design for patients diagnosed with Nephroblastoma at KNH, and by extension, other tertiary public health care facilities in Kenya. Patients and methods: The records of 140 WT patients, aged less than 16 years, who were treated in Kenyatta National Hospital, Kenya, during the period from January 1997 to December 2008 were reviewed. The management protocol followed the scheme of the US National Wilms Tumor Study Group (NSWTG). Results: Thirteen cases (38.2%) were diagnosed as stage I, 4 (11.8%) as stage II, 13 (38.2%) as stage III and 2 (5.9%) as stage IV. Eight cases had bilateral disease (stage V). Four-year overall survival (OS) and event free survival (EFS) rates were xx% and xx%, respectively. Univariate analysis by Log-rank test revealed statistically significant associations between OS and nodal status (p-value < 0.01), manifestation of gross hematuria (p-value 0.02), and tumor size of 10 centimeters or more (p-value 0.02). Multivariate analysis found only the nodal status to be independently associated with OS at a Hazard Ratio of 16.6 (p-value < 0.01). Eight of 13 stage I cases and 6/13 stage III cases had relapsed, with two-year post-relapse survival of 42.8%. Significantly poorer outcome was found in cases with early relapse within 200 days after enrollment (p-value 0.02). Conclusion: Childhood Wilms' tumor presents late in our setting with its consequent management challenges. The need to educate the populace and the primary healthcare providers on the benefits of early diagnosis and treatment of this condition cannot be overemphasized. Large tumor size and gross hematuria were associated with risk of a poorer outcome.