Lipiodol brain embolism during hepatic transcatheter arterial chemoembolization

Abstract

Sir, Transcatheter arterial chemoembolization (TACE) for advanced hepatocellular carcinoma, i.e., the injection of a mixture of iodized oil (lipiodol) and a chemotherapeutic via the tumor feeding artery, hitherto has been complicated by lipiodol brain embolism (LBE) in a few cases [1–5]. Only two of these patients had repeat MRI that documented extensive reversion of the initial DWI abnormalities [5]. We present a case of TACE-associated LBE that demonstrated near-complete resolution of DWI lesions suggesting cytotoxic edema. During his fourth course of TACE, a 71-year-old man with advanced hepatocellular carcinoma presented a modest monoparesis of the right arm along with transient dysarthria and dizziness. CT scan immediately after the procedure showed multiple infraand supratentorial brain lesions consistent with the deposition of iodized oil. Three hours later, MRI disclosed multiple nonenhancing cortical and subcortical hyperintense lesions, which represented areas of decreased diffusion on DWI. In particular, the magnitude of ADC decrease in the lesioned areas ranged roughly between 15 and 25% compared to contralateral normal-appearing brain tissue. The patient recovered completely over the following 48 h. A repeat MRI after 1 month documented near-complete resolution of the lesions (Fig. 1). Hyperintense DWI lesions that exhibit low ADC values are typically suggestive of cytotoxic edema in a setting of acute brain ischemia. Although ischemia-associated DWI hyperintensities can be reversible to a certain extent, it is remarkable that in a setting of both cerebral fat [6, 7] and LBE post-TACE [5], widespread areas of presumed cytotoxic edema have been reported not to evolve to complete infarction (established T2 or FLAIR lesions). To the best of our knowledge, such an extensive reversal of DWI abnormalities has not been reported in acute ischemic stroke. A possible explanation would be that what looks like cytotoxic edema in DWI is actually an overestimation of the true extent of the lesion or a cytotoxic edema ‘‘mimic.’’ A MRI spectroscopy study [6] has suggested that soon after fat embolism, areas of restricted diffusion may simply reflect the presence of large quantities of ‘‘sluggish’’ lipids compared to water. However, secondary ischemia [6] or a neurotoxic effect of free fatty acids [7] may occur later if the embolic material is not rapidly cleared from the brain circulation. This would also explain the ‘‘preference’’ of the lipid emboli and of the associated permanent lesions for the watershed areas of the brain where a slow blood flow cannot be compensated quickly and adequately by collaterals. In our patient, fast clearance of the lipid burden may account for the rapid clinical recovery and the reversion of DWI abnormalities. Concerning the etiology of brain embolism in patients undergoing TACE, right-to-left shunt (either via a patent foramen ovale or intrapulmonary arteriovenous shunts), shunt from the inferior phrenic artery to the pulmonary vasculature, and an increased dose of infused lipiodol are commonly advocated [1–4]. However, the passage of oil T. Karapanayiotides (&) G. Georgiadis Department of Neurology, Hippokrateion Hospital, 50, Aigaiou Str., 551 33 Thessaloniki, Greece e-mail: theonen@ath.forthnet.gr; tkarapanayiotides@yahoo.gr