Background Multifocal intrahepatic cholangiocarcinoma (ICC) has traditionally been treated with surgical resection when amenable. Intra-arterial therapy (IAT) for multifocal ICC has not been directly compared with surgical resection. Methods A single-center, retrospective review of consecutive patients treated for multifocal ICC was conducted. Patients with distant metastases or treatment with systemic chemotherapy alone were excluded. Patients were divided into two groups: surgical resection versus IAT; IAT included transarterial chemoembolization (TACE), transarterial radioembolization (TARE), and hepatic arterial infusion (HAI) pump therapy. Subjects were also analyzed by surgical resection, TACE, and HAI pump therapy. Results Overall, 116 patients with multifocal ICC were studied, 57 in the surgical resection group and 59 in the IAT group (TACE = 41, HAI pump = 16, TARE = 2). The IAT group was characterized by a higher incidence of bilobar disease (88.1% vs. 47.4%, p  < 0.001), larger tumors (median 10.6 vs. 7.5 cm, p  = 0.004), higher incidence of macrovascular invasion (44.1% vs. 24.6%, p  = 0.027), and higher rate of nodal metastases (57.6% vs. 28.6%, p  = 0.002). Median overall survival for surgical resection was 20 months versus 16 months for IAT ( p  = 0.627). Multivariate analysis found that macrovascular invasion hazard ratio (HR) 2.52, 95% confidence interval (CI) 1.56–4.09 and non-receipt of systemic chemotherapy (HR 3.81, 95% CI 2.23–6.52) were independent poor prognostic risk factors. Surgical resection was not associated with a survival advantage over IAT on multivariate analysis ( p  = 0.242). Conclusion Despite selection bias for use of surgical resection compared with IAT, no survival advantage was conferred in the treatment of multifocal ICC.