Iodized oil (lipiodol, an iodized fatty acid ethyl ester, fatty acid derived from poppy seed oil, Andre Guebet, Aulnay-sous-Bios, France) has been widely used as a basis of lymphangiographic dyes and for transcatheter arterial chemoembolization (TAE) with carcinostatics in the treatment of hepatocellular carcinoma. Various complications arising during lymphangiography or TAE have been reported, including fever, pain, and acute hepatic failure [1-3]. Other more serious complications have also been reported such as pulmonary embolism by the lipiodol especially after lymphangiography . However, the incidence of pulmonary embolism induced by lipiodol during TAE is infrequent. Chung and his colleagues  reported that respiratory symptoms occurred in 3% (11 in 336 patients) of patients undergoing TAE, and that pulmonary embolism caused by lipiodol, which appears as bilateral diffuse pulmonary parenchymal infiltration, occurred in 1.7% of patients. In these cases, the initial chest radiograph was normal and the onset of the respiratory symptoms was delayed for 2-4 days after TAE. Here we present a case of early onset pulmonary embolism caused by lipiodol, which was detected during TAE.
TAE was planned for a 7-year-old child (body weight 20 kg), using a mixture of lipiodol and a carcinostatic (pirarubicin hydrochloride) for selective injection into the hypervascular area of the hepatoblastoma. The tumour was located in the right lobe of the liver and had invaded the inferior vena cava and right atrium. Atrial premature beats were recorded, which were probably due to a direct mechanical stimulation by the extension of the tumour into the right atrium. Anaesthesia was induced with midazolam (2.5 mg), nitrous oxide, sevoflurane, and vecuronium, and maintained with 50% oxygen and isoflurane without nitrous oxide. The patient was carefully monitored using standard monitoring as well as direct radial arterial monitoring because haemodynamic instability caused by pulmonary embolism was anticipated due to the extension of the tumour to the right atrium.
The initial embolization of the right and middle hepatic arteries was performed with gelatin sponge particles (Gelforme, Upjohn, Kalamazoo, MI, USA) during angiography. During TAE for the middle hepatic artery, arterial oxyhaemoglobin saturation (SpO2) suddenly decreased from 100 to 90% together with the appearance of frequent atrial premature beats. TAE was immediately stopped and the inspired oxygen was increased to 100%. By this stage, a mixture of lipiodol 14 mL and carcinostatic (pirarubicin hydrochloride) 50 mg had been given. Chest fluoroscopy indicated the spread of lipiodol bilaterally, especially into the middle lung zones, and strongly indicated lipiodol-induced pulmonary embolism. Anticoagulation therapy was instituted, consisting of a bolus injection of 2000 units (U) heparin followed by a continuous infusion of 200 U h−1 heparin to maintain partial the thromboplastin time at about 50 s. Analysis of arterial blood gases during 100% O2 breathing showed the arterial partial pressure of O2 (PaO2) to be 26 kPa and the PaCO2 7.5 kPa. The arterial to end-tidal CO2 partial pressure difference was large (2.13 kPa). However, signs of right heart failure were not seen.
After the completion of TAE, the patient was transferred to the intensive care unit and treated with mechanical ventilation [positive end-expiratory pressure (PEEP) 4-5 cmH2O, pressure support ventilation]. The patient was treated with heparin 200 U h−1, nitroglycerine 0.5-1.5 μg kg−1 min−1, furosemide, and methylprednisolone 10 mg kg−1 for 5 days.
The chest radiographic abnormality deteriorated to diffuse bilateral pulmonary parenchymal infiltrates, mainly in the middle fields of the lungs, compatible with the adult respiratory distress syndrome (Fig. 1). Computed tomography of the chest immediately after TAE showed accumulation of lipiodol in the middle lung fields (Fig. 2). Arterial hypoxaemia and arterial to end-tidal PCO2 difference recovered to normal values 1 day after TAE. However, the pulmonary infiltrates of the chest radiograph persisted although the abnormality gradually improved to full clearance at 7 days after TAE (Fig. 3).
Respiratory symptoms caused by lipiodol are infrequent and develop in less than 2% of patients undergoing lymphangiography . More severe complications, e.g. pulmonary infarction, have also been reported . Among the complications associated with lipiodol, pulmonary embolism is the most serious. It occurs in 2-4% of patients undergoing TAE [1,2,5].
Respiratory distress usually occurs several days after lymphangiography. The mechanism of delayedonset pulmonary oedema is thought to be due to chemical injury caused by fatty acid components . Microemboli pass via lymphatic-venous anastomosis, and are gradually broken down by esterases of either lung or leukocyte origin. The relatively long time required for the rise of free fatty acid concentrations to toxic levels probably explains the delayed onset of pulmonary oedema in these patients .
Samejima and his colleagues  reported a case with acute onset pulmonary distress caused by lipiodol after intra-arterial infusion chemotherapy using lipiodol-adriamycin emulsion. In that report, severe dyspnoea and cyanosis occurred at 30 min after infusion, and the patient was diagnosed as having pulmonary oedema due to adriamycin .
In our case one or more mechanisms may have participated in the development of this serious complication. First, pulmonary embolism caused by lipiodol may have occurred through arterio-venous shunts in the hepatoblastoma, resulting in multiple mechanical occlusions of the pulmonary artery. Subsequently this may have caused the sudden fall in SpO2, the increased difference between PaCO2 and PETCO2, and the atrial premature beats. Second, delayed-onset pulmonary oedema due to chemical injury caused by free fatty acids, as described above may have contributed. Third, the carcinostatic agent may have induced further lung injury. The later pulmonary complication, especially reduced oxygenation due to interstitial pneumonitis induced by carcinostatics is known to cause a delayed onset pulmonary oedema [9,10].
The suggested treatment of this complication is similar to that used for the fat embolism syndrome. It can be divided into general supportive and 'specific' therapy. The general supportive treatment includes respiratory support such as oxygen therapy, mechanical ventilation, PEEP, and parenteral fluids. 'Specific' therapy such as high-dose corticosteroids, heparin, ethanol, hypertonic glucose and insulin have also been advocated . However, there are no data demonstrating the effectiveness of these therapeutic modalities [12,13].
It is obviously important to reduce the risk of pulmonary oil embolism during TAE therefore Chung and his colleagues  recommended that the dose of lipiodol should be < 20 mL (0.25 mL kg−1) in adults. They also suggested that it should be determined whether or not the patient had a significant arteriovenous hepatic shunt before the treatment was performed . They demonstrated that the use of > 0.3 mL kg−1 of lipiodol was associated with the development of pulmonary oil embolism in 43% of their patients . If the volume of the tumour treated by TAE is too large, the sandwich technique should be used with alternate administration of the lipiodol/carcinostatic emulsion and gelatin sponge particles . In our case, we also used this sandwich method with 14 mL (0.7 mL kg−1) of lipiodol. The volume of lipiodol was greater than the dose recommended by Chung and his colleagues . Although we carefully examined the arteriogram for arterio-venous shunt, no abnormality could be detected. Samejima and his colleagues  also reported pulmonary complications without clear evidence of any arterio-venous shunt using arteriography . In experimental studies in dogs, Kishi and his colleagues  demonstrated that lipiodol infusion into the hepatic artery resulted in dose-dependent increase in the circulating levels of the agent. Furthermore, they reported the potential development of embolism in dogs when the lipiodol droplets pass via sinusoids and pulmonary capillaries into the systemic circulation, although cerebral infarction has not yet been reported. Therefore, pulmonary embolism caused by lipiodol can be difficult to control in patients who receive a high-dose lipiodol, especially if they have intrahepatic arterio-venous shunts, even when the shunt cannot be identified by arteriography. Thus, early signs of pulmonary embolism include an unexpected reduction in SpO2 during TAE. Furthermore, care should also be taken because of the potential development of late-onset lung injury after pulmonary embolism, such as chemical injury caused by the lipiodol and interstitial pneumonitis caused by high dose of carcinostatics.
In conclusion, pulmonary embolism caused by lipiodol during TAE is infrequent, but such a complication could be fatal. Understanding the risk of pulmonary embolism in patients receiving lipiodol, especially high dose, during and after TAE, and lateonset pulmonary injury after TAE is important and a close follow-up for several days after TAE is advisable.
The authors thank F. G. Issa, MD, PhD (Word-Medex, Sydney, Australia) for the careful reading and editing of the manuscript.
1 Sakamoto I, Aso N, Nagaoki K et al.
Complications associated with transcatheter arterial embolization for hepatic tumors. Radiographics
2 Chung JW, Park JH, Han JK et al.
Hepatic tumors: predisposing factors for complications of transcatheter oily chemoembolization. Radiology
3 Yamada R, Sato M, Kawabata M, Nakatsuka H, Nakamura K, Takashima S. Hepatic artery embolization in 120 patients with unresectable hepatoma. Radiology
4 Silvestri RC, Huseby JS, Rughani I, Thorning D, Culver BH. Respiratory distress syndrome from lymphangiography contrast medium. Am Rev Resp Dis
5 Chung JW, Park JH, Im JG, Kim SK, Park YB, Han MC. Pulmonary oil embolism after transcatheter oily chemoembolization of hepatocellular carcinoma. Radiology
6 Dolan PA. Lymphography: complications encountered in 522 examinations. Radiology
7 Takahashi M, Abrams HL. Arborizing pulmonary embolization following lymphangiography. Report of three cases and an experimental study. Radiology
8 Goff AM, Gaensler EA. Respiratory distress syndrome following lymphangiography. Respiration
9 Samejima M, Tamura S, Kodama T et al.
Pulmonary complication following intra-arterial infusion of lipiodol-adriamycin emulsion for hepatocellular carcinoma, report of a case. Nippon Igaku Hoshasen Gakkai Zasshi Nippon Acta Radiol
10 Kudo S, Wright KC, Chuang VP, Wallace S, Mir S, Bechtel W. Experimental evaluation of intraarterial occlusion-infusion chemotherapy. Am J Roentgenol
11 Richards RR. Fat embolism syndrome. Can J Surg
12 Kallenbach J, Lewis M, Zaltzman M, Feldman C, Orford A, Zwi S. 'Low-dose' corticosteroid prophylaxis against fat embolism. J Trauma Injury Infect Crit Care
13 Fulde GW, Harrison P. Fat embolism - a review. Arch Emerg Med
14 Kishi K, Sonomura T, Satoh M et al.
Acute toxicity of lipiodol infusion into the hepatic arteries of dogs. Invest Radiol