Feasibility of using CT volume as a predictor of specimen weight in a subgroup of patients with low risk Wilms tumors registered on COG Study AREN03B2: implications for central venous catheter placement

Abstract

Objective: Patients with stage I Wilms tumor, age ≤ 2 years, tumor ≤ 550 g may not require therapy beyond nephrectomy. This study's aims were to determine: (1) if a linear relationship exists between tumor weight and computed tomography (CT) estimated volume; (2) describe the accuracy of a slope-intercept equation in estimating weight; and (3) determine the potential impact of weight estimation on port placement decisions.

Materials and methods: Tumor weight and port placement information were abstracted from 105 patients, age ≤ 2 years, with tumors ± 550 g, enrolled in COG AREN03B2. One radiologist estimated tumor size from CT scan. Prolate ellipse volume (PEV) was calculated, linear regression performed, slope-intercept equation calculated, equation estimated weight determined, and potential impact of the on port placement evaluated.

Results: A strong relationship exists between PEV and weight (R(2) = 0.87). The slope-intercept equation for weight was: weight = 1.04(PEV) + 58.75. Overall median relative error for the equation was 0.9%, and -3% in tumors weighing 350-750 g. Fifty-five ports were placed, 29 in patients with tumor weight ≤ 550 g, and six not placed in patients with tumor weight > 550 g.

Conclusions: The relationship between PEV and weight produced a reliable weight prediction equation. Preoperative consideration of specimen weight may diminish the number of ports placed in this population.

Keywords: Central venous catheter; Computed tomography; Nephrectomy; Nephroblastoma; Wilms tumor.