Abstract Background Isolated pelvic perfusion (IPP) can be used to treat unresectable melanoma metastases of the pelvis. IPP can be performed either by surgical or percutaneous approaches, using different balloon catheters. The aim of this study was to examine whether the surgical and percutaneous approaches were comparable with respect to tumor drug exposure in the pelvis. Methods A pharmacokinetic study was performed in 5 melanoma patients treated with surgical IPP and five with percutaneous IPP. Both groups received melphalan at the dose of 30 mg/m². Melphalan pharmacokinetic analyses were performed and the main parameter used to evaluate pelvic tumor drug-exposure was the ratio of areas under the melphalan plasma concentration curves in the pelvis and the systemic compartment, during the perfusion time (AUC_{0 to 20}). Non-parametric Mann–Whitney tests were employed for statistical comparisons. Results The median and interquartile range (IQR) values of the ratios between melphalan AUC_{0 to 20} in pelvic and systemic compartments were 7.9 (IQR 7.2 to 9.9) and 5 (IQR 4 to 7.9) for surgical and percutaneous IPPs, respectively (p = 0.209). Conclusions Tumor exposure to drug using these two methods did not statistically differ and both methods, therefore, can be adopted interchangeably, utilizing a perfusion blood flow rate of approximately 120 ml/min. The small sample size is a limitation of this study but our preliminary results can be used to calculate the effect size of a larger trial. Trial Registration Clinical Trials.gov Identifier NCT01920516; date of trial registration: August 6, 2013