976 Clinical report
Risk factors for the development of pulmonary oil embolism after transcatheter arterial chemoembolization of hepatic tumors Geng-Chin Wua, Edward D. Chanh,i,j, Yu-Ching Choub, Chih-Yung Yuc, Tsai-Yuan Hsiehd, Chung-Bao Hsiehe, Chih-Feng Chianf, Fu-Chang Kea, Yu-Ling Daif and Wen-Lin Sub,f,g Pulmonary oil embolism (POE) is a rare fatal complication after transcatheter arterial embolization (TAE) and transcatheter arterial chemoembolization (TACE). As risk factors have not been clearly delineated, the aim of the present study was to identify the risk factors for development of POE after TACE. A retrospective analysis was carried out on patients with unresectable hepatocellular carcinoma who received TAE or TACE at the Tri-Service General Hospital (Taiwan) between January 2005 and December 2008. The diagnosis of TAE-induced or TACE-induced POE was based on development of respiratory signs and symptoms relatively soon after the procedure, as well as based on characteristic radiographic findings. Of the 219 enrolled patients in this study, 20 were diagnosed with POE after TAE or TACE. On univariate logistic regression analysis, patients developing POE were found to be older (67.95±15.95 vs. 61.44±12.59 years, P = 0.033), with a lower serum albumin level (3.25±0.58 vs. 3.62±0.57 g/dl, P = 0.009), a higher grade of liver cirrhosis as classified on the basis of Child’s criteria (P < 0.006), a larger tumor size (8.55±4.52 vs. 4.78±3.97 cm in diameter, P < 0.001), a higher lipioidol dose (22.35±11.01 vs. 13.69±7.66 ml, P = 0.003), and a higher doxorubicin dose (50.27±7.05 vs. 40.75±13.61 mg, P < 0.001). Following multivariate logistic regression analysis, only lipiodol dose was found to be a significant risk factor for POE (odds ratio = 1.133, 95% confidence interval: 1.004, 1.279; P = 0.044). The receiver operator characteristic curve cutoff
Introduction Hepatocellular carcinoma (HCC) is one of the most common malignant tumors in man and one of the deadliest [1]. Transcatheter arterial embolization (TAE) and transcatheter arterial chemoembolization (TACE) are relatively effective palliative treatments for unresectable HCC. Although TACE has been shown to increase survival [2,3], a recent meta-analysis found that the benefit of TACE was still limited [4]. Ethiodized oil (lipiodol), comprising iodine combined with ethyl esters of fatty acids of poppyseed oil, has been widely used as an embolic agent and a carrier of chemotherapeutic agents in the treatment of HCC [5,6]. Various complications arising during TACE have been reported, including acute hepatic failure, liver abscess, intrahepatic biloma formation, liver c 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins 0959-4973
point for lipiodol dose level was 14.5 ml, with a sensitivity of 80% and a specificity of 66.3%. In conclusion, the lipiodol dose could be considered as a predictive factor for POE after TAE or TACE in hepatic malignant tumor patients. On the basis of this retrospective study, the safe lipiodol dose to minimize the risk for POE is 14.5 ml or lower; however, larger, prospective studies are needed to determine the optimally safe and yet efficacious c 2014 Wolters dose. Anti-Cancer Drugs 25:976–981 Kluwer Health | Lippincott Williams & Wilkins. Anti-Cancer Drugs 2014, 25:976–981 Keywords: hepatocellular carcinoma, pulmonary oil embolism, transcatheter arterial chemoembolization a Department of Internal Medicine, Division of Pulmonary Medicine, Taoyuan Armed Forces General Hospital, Lungtan, Taoyuan, bSchool of Public Health , National Defense Medical Center, cDepartment of Radiology, dDepartment of Internal Medicine, Division of Gastroenterology, eDepartment of Surgery, Division of Organ Transplantation, fDepartment of Medicine, Division of Pulmonary and Critical Care Medicine, Tri-Service General Hospital, Nei-Hu, Taipei, gDepartment of Medicine, Division of Pulmonary and Critical Care Medicine, Taipei Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation, New Taipei, Taiwan, Republic of China, hDivision of Pulmonary Sciences and Critical Care Medicine, Department of Medicine, University of Colorado Denver at Anschutz Medical Center, iDenver Veterans Affairs Medical Center and jDepartment of Medicine, Division of the Mycobacterial and Respiratory Infections, National Jewish Health, Denver, Colorado, USA
Correspondence to Wen-Lin Su, MD, PhD, Department of Medicine, Division of Pulmonary and Critical Care Medicine, Taipei Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation, No.289, Jianguo Rd, Xindian Dist., New Taipei City 231, Taiwan, Republic of China Tel: + 886 2 8792 7233; fax: + 886 2 6600 8692; e-mail: [email protected] Received 2 November 2013 Revised form accepted 8 March 2014
infarction, multiple intrahepatic aneurysms, gallbladder infarction, cholecystitis, splenic infarction, gastrointestinal mucosal lesions, pulmonary oil embolism (POE), tumor rupture, and variceal bleeding [7–10]. POE is an extremely rare but serious complication after TACE [7,11]. Sakamoto et al. [7] reported that POE occurred with a frequency of 0.17% after TACE. The amount of lipiodol infused and the presence of arteriovenous shunting in the tumor have been shown to increase the risk for POE [11–13]. Chung et al. [11] recommended that the maximum safe dose of lipiodol is 15–20 ml/patient (B0.25 ml/kg). Nevertheless, patients without arteriovenous shunts developing POE while receiving recommended doses of lipiodol for TACE have been reported [14]. DOI: 10.1097/CAD.0000000000000113
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Risk factors for development of POE Wu et al. 977
Identifying additional risk factors for the development of POE after TAE/TACE can be clinically useful for, perhaps, instituting preventive measures such as dose reduction in high risk individuals. The present study aimed to evaluate the clinical feature of patients with POE and to identify the risk factors for its development.
Materials and methods Patients
In this single-center, retrospective study, consecutive patients who underwent TAE or TACE between January 2005 and December 2008 at Tri-Service General Hospital for unresectable HCC or metastatic cancer of the liver were analyzed for the development of POE. The study was approved by the Internal Review Board at Tri-Service General Hospital. The diagnosis of HCC was confirmed through histologic and cytologic findings or abdominal multidetector computed tomography (MDCT) findings consistent with HCC [15,16]. Cases of metastatic cancer to the liver were confirmed histopathologically.
by the development respiratory signs and symptoms (cough, dyspnea, hemoptysis, and tachypnea), reduced gas exchange (decreased oxygen saturation on arterial blood gas measurements), and the presence of ill-defined, high-density reticular opacities on chest radiographs (Fig. 1) and/or chest computed tomography (CT) scans (Fig. 2) within 7 days after administration of TAE or TACE. As the clinical manifestations of POE are nonspecific and may mimic other diseases such as heart failure, infectious pneumonia, chronic airway or interstitial lung disease, and transfusion-related acute lung injury, we also stipulated that patients classified as having POE have normal left ventricle ejection fraction on echocardiography, negative sputum and blood cultures, no clinical evidence of fluid overload, no history of acute or chronic lung diseases such as asthma, chronic obstructive pulmonary disease, and interstitial lung disease, no recent transfusion of any blood products (before developing respiratory symptoms after TAE or TACE), and no infiltration on chest radiographs before TAE or TACE. Statistical analysis
Chemoembolization
All participating radiologists had prior experience with lipiodol chemoembolization. Hepatic angiography and MDCT were routinely performed to delineate the blood supply, the arteriovenous shunt, and the size and extent of the HCC. TAE was initiated with the infusion of iodized oil (lipiodol; Guerbet, Aulnay-sous-Bois, France) through the proper hepatic artery, left hepatic artery, or right hepatic artery, which supplied the tumors. For TACE, an emulsion consisting of lipiodol and doxorubicin hydrochloride (adriamycin; Pfizer Italia S.R.L., Milano, Italy) was infused. The amount of iodized oil and doxorubicin hydrochloride used ranged from 4 to 50 ml and 6 to 60 mg, respectively. These dosages were determined on the basis of tumor size and vascularity.
Patients were categorized into groups with and without POE following TAE or TACE. As most patients had undergone TAE or TACE more than once, the data were obtained following the first administration of TAE or TACE. Continuous and categorical variables are presented as mean±SD and number (percentage), respectively. Fig. 1
Embolization was subsequently performed with gelatinsponge particles (Spongostan standard; Johnson & Johnson Medical Pty. Ltd, Gauteng, South Africa) until flow in the hepatic artery was sluggish. The goal was to decrease arterial flow without causing total obstruction. Data collection
The following data were extracted from patient medical records: demographic characteristics (including age and sex), body weight, results from standard blood analysis (including white blood cell count and C-reactive protein, aspartate aminotransferase, alanine aminotransferase, total bilirubin, albumin, alkaline phosphatase, and g-glutamyl transpeptidase levels), etiology of hepatic tumor, Child’s classification of the liver before treatment, tumor size, method of embolization (TAE or TACE), and dosages of lipiodol (ml) and doxorubicin (mg). Definition and diagnosis
Chest radiographs obtained during admission were analyzed by a chest radiologist and a pulmonologist. POE was defined
Chest radiograph showing high-density ill-defined reticular opacities of both lungs.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
978 Anti-Cancer Drugs 2014, Vol 25 No 8
Fig. 2
Fig. 3
ROC curve 1.0
0.8
Sensitivity
0.6
0.4
0.2
0.0 0.0
0.2
0.4
0.6
0.8
1.0
1−Specificity Noncontrast CT scan showing linear high-density shadows (black arrow) in both lung fields, indicating lipiodol retention. CT, computed tomography.
P-values were determined by t-test or w2/Fisher’s Exact Test, by Fisher’s Exact Test, or by w2-test. Univariate logistic regression analysis and multivariate logistic regression analysis were carried out on continuous variables to identify factors of POE associated with TAE or TACE. A receiver operator characteristic curve (Fig. 3) was plotted to determine the maximum safe dose of lipiodol. Analyses were carried out using the SPSS 17.0 software package (SPSS Institute Inc., Chicago, Illinois, USA). All statistical assessments were two-sided, using a significance level of 0.05.
Receiver operator characteristic (ROC) curve for lipiodol dose level.
Fig. 4
Patients with unresectable hepatocellular carcinoma who underwent TAE or TACE (219 elevated)
Patients had respiratory symptoms after TAE or TACE (n = 25)
Results Of the 219 patients who underwent TAE or TACE, 205 had HCC and 14 had metastatic cancer to the liver. Respiratory signs and symptoms, including cough, hemoptysis, dyspnea, and/or tachypnea, developed within 1 h to 6 days after TACE and TAE. Twenty (9.1%) of the 219 patients met the criteria of having POE after the procedure (Fig. 4). Whereas five additional patients did have respiratory symptoms after TAE or TACE, they did not show high-density reticular opacities in the lungs on imaging and thus were not defined as having POE. All 219 patients underwent angiography and MDCT. Interestingly, these imaging studies did not reveal evidence of arteriovenous shunting in any of the patients. Table 1 lists the demographic characteristics and laboratory values of the 219 patients. The cause of HCC was almost always chronic hepatitis B or chronic
No infiltration on chest X-ray (n = 5)
Patients with pulmonary oil embolism after TAE or TACE (n = 20)
Patients without pulmonary oil embolism after TAE or TACE (n = 199)
Patients who received TAE or TACE and who had clinical symptoms and radiographic abnormalities consistent with pulmonary oil embolism. TACE, transcatheter arterial chemoembolization; TAE, transcatheter arterial embolization.
hepatitis C infections in those with or without POE. Patients who developed POE after TAE or TACE procedures were found to be older (67.95±15.95 vs.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Risk factors for development of POE Wu et al. 979
Comparison of demographics and clinical features between pulmonary embolism and non-pulmonary embolisma
Table 2 Univariate logistic regression analysis of pulmonary embolism rate
Table 1
Variables
PE (n = 20) [N (%)]
Non-PE (n = 199) [N (%)]
P-valueb d
Sex 1.000 Male 13 (65.0) 128 (64.3) Female 7 (35.0) 71 (35.7) Age 67.95±15.95 61.44±12.59 0.033 Body weight (kg) 61.6±14.2 64.45±10.81 0.277 Laboratory data WBC (/ml) 6719.0±3296.0 5298.92±2170.38 0.074 CRP (mg/dl) 15.27±7.90 11.52±10.58 0.363 AST (U/l) 105.55±100.85 76.85±75.18 0.117 ALT (U/l) 71.8±68.5 67.15±63.26 0.756 T.bil (mg/dl) 1.56±1.86 1.18±1.27 0.223 Alb (g/dl) 3.25±0.58 3.62±0.57 0.009 ALP (U/l) 272.77±299.10 124.19±92.56 0.100 g-GT (U/l) 198.7±158.9 137.71±136.56 0.265 Etiology 0.960c Alcohol 0 (0) 6 (3.0) HBV 11 (55.0) 108 (54.3) HCV 8 (40.0) 67 (33.7) META 1 (5.0) 13 (6.5) Unknown 0 5(2.5) Child’s 0.006c classification of liver cirrhosis A 10 (50.0) 161 (80.9) B 9 (45.0) 36 (18.1) C 0 1 (0.5) N 1 (5.0) 1 (0.5) Tumor size (cm) 8.55±4.52 4.78±3.97 < 0.001 Lipiodol dose 22.35±11.01 13.69±7.66 0.003 (ml) Adriamycin (mg) 50.27±7.05 (n = 15) 40.75±13.61 (n = 148) < 0.001
Alb, albumin; ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CRP, C-reactive protein; HBV, hepatitis B virus; HCV, hepatitis C virus; META, metastatic; N, non-cirrhosis of liver; T.bil, total bilirubin; WBC, white blood cell; g-GT, g-glutamyl transpeptidase. a Continuous and categorical variables are presented as mean±SD and number (%), respectively. b P-value by t-test or w2/Fisher’s exact test. c Fisher’s exact test. d 2 w -test.
61.44±12.59 years, P = 0.033), with lower serum albumin levels (3.25±0.58 vs. 3.62±0.57 g/dl, P < 0.01), more advanced stage of cirrhosis based on Child’s classification (P < 0.006), and larger tumor size (8.55±4.52 vs. 4.78± 3.97 cm in diameter, P < 0.001). Patients who developed POE also were found to have received higher doses of lipiodol (22.35±11.01 vs. 13.69±7.66 ml, P = 0.003) and doxorubicin (50.27±7.05 vs. 40.75±13.61 mg, P < 0.001).
Variables Sex (M/F) Age Body weight (kg) Laboratory data WBC (/ml) CRP (mg/dl) AST (U/l) ALT (U/l) T.bil (mg/dl) Alb (g/dl) ALP (U/l) g-GT (U/l) Etiology Others (alcohol + META + Unknown) HBV HCV Non-HBV (HCV + alcohol + META + unknown) HBV Child’s classification of liver cirrhosis A+N B+C Tumor size (cm) Lipiodol dose (ml) Adriamycin (mg)
Odds ratio
95% Confidence interval
1.030 1.047 0.976
0.393, 2.700 1.003, 1.092 0.935, 1.019
0.952 0.034 0.277
1.002 1.038 1.003 1.001 1.159 0.328 1.005 1.002
1.0015, 1.0024 0.960, 1.122 0.999, 1.007 0.994, 1.008 0.905, 1.486 0.139, 0.776 1.001, 1.008 0.998, 1.007
0.012 0.355 0.140 0.755 0.242 0.011 0.009 0.274
1.0
P-value
Reference
2.444 2.866 1.0
0.301, 19.849 0.340, 24.126 Reference
0.403 0.333
1.030
0.409, 2.595
0.950
1.0 3.582 1.167 1.098 1.070
Reference 1.385, 9.268 1.07, 1.273 1.048, 1.151 1.014, 1.13
0.009 < 0.001 < 0.001 0.013
Alb, albumin; ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CRP, C-reactive protein; HBV, hepatitis B virus; HCV, hepatitis C virus; META, metastatic; N, non-cirrhosis of liver; T.bil, total bilirubin; WBC, white blood cell; g-GT, g-glutamyl transpeptidase. P < 0.05.
Table 3 Multivariate logistic regression analysis of pulmonary embolisma Variables
Odds ratio
Age 1.039 Laboratory data WBC (/ml) 1.0001 Alb (g/dl) 2.160 ALP (U/l) 1.003 Child’s classification of liver cirrhosis A+N 1.0 B+C 2.462 Tumor size (cm) 1.028 Lipiodol dose (ml) 1.133 Adriamycin (mg) 1.031
95% Confidence interval
P-value
0.964, 1.119
0.316
0.9998, 1.0004 0.334, 13.979 0.998, 1.007
0.613 0.419 0.222
Reference 0.247, 24.552 0.862, 1.226 1.004, 1.279 0.950, 1.118
0.442 0.759 0.044 0.470
Alb, albumin; ALP, alkaline phosphatase; WBC, white blood cell. The statistically significantly factors (P < 0.05), including the continuous and categorical variables from Table 2. a
Table 2 shows the results of univariate logistic regression analysis. Increasing age, white blood cell count, alkaline phosphatase level, tumor size, lipiodol dose, and doxorubicin dose, as well as severity of cirrhosis, were associated with increased risk for POE. In contrast, increased serum albumin levels were associated with lower risk for POE. Following multivariate logistic regression analysis, only the dose of lipiodol was found to be an independent risk factor for the development of POE (Table 3). On the basis of a receiver operator characteristic curve for lipiodol dose and POE, lipiodol dose higher than 14.5 ml was found to diagnose POE with a sensitivity of 80% and a
specificity of 66.3%. The overall mortality rate for patients with POE was 25%; the five patients who died had developed multiorgan failure, including respiratory failure, hepatic failure, and oliguria with acute renal failure.
Discussion In this retrospective analysis of 219 patients who underwent TAE or TACE for unresectable malignant tumor of the liver, we found, on multivariate regression analysis, that the lipiodol dose was the primary risk factor for POE. It makes intuitive sense that the amount of lipiodol
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
980 Anti-Cancer Drugs 2014, Vol 25 No 8
infused is a risk factor for the development of POE, as has been suggested by others [11–13]. Indeed, Chung et al. [11] recommended that the maximum safe dose of lipiodol is 15–20 ml/patient (B0.25 ml/kg). Our finding that a lipiodol dose greater than 14.5 ml is associated with the development of POE would lend credence to this recommendation. Intratumor arteriovenous shunting is considered to be a risk factor for POE. MDCT is an important medical imaging examination technology to detect the hepatic arteriovenous shunt associated with HCC [17]. In the present study, all patients underwent both angiography and MDCT, and arteriovenous shunts were not detected in any patient, a finding consistent with those of previous studies [11–14,18]. The inability to detect arteriovenous shunts may be related to the decreased sensitivity of angiography to detect microvascular shunts and thus the possibility that the presence of such shunts predispose to POE is not ruled out. The technetium-99m-macroaggregated albumin test may be more sensitive in detecting lung shunting in HCC [19,20]. Ho et al. [21] demonstrated that lung shunting was correlated with tumor size and tumor vascularity in patients with HCC. Although our univariate regression analysis did show an increased risk for POE based on tumor size, multivariate analysis failed to show tumor size as an independent risk factor for POE. Intra-arterial infusion chemotherapy with doxorubicin is reported to be complicated by thrombosis of regional blood vessels and local toxicity [22]. Although the results of univariate logistic regression analysis in our study show that the dose of doxorubicin is a risk factor for POE, multivariate analysis failed to demonstrate this. The route by which oil droplets injected into the liver accumulate in the pulmonary vasculature is not precisely known. Lipiodol infusion into the hepatic artery of dogs resulted in dose-dependent POE. These observations substantiated embolization of oil droplets through the hepatic sinusoids to pulmonary capillaries [23]. Further, POE was probably associated with hepatic vein invasion or communication between the tumor feeding artery and the pulmonary vein that might occur as a result of tumor invasion into the pleura or diaphragm [24,25]. Thus, it is reasonable to posit that the pathogenesis of POE likely involves migration of oil droplets from the hepatic sinusoids into the pulmonary capillaries through microvascular shunts, hepatic vein invasion, or tumor invasion into the diaphragm or pleural space. The mechanism by which lipiodol itself causes pulmonary toxicity is not well understood, although direct chemical injury has been postulated, as free fatty acids injected intravenously result in the rapid onset of severe lung injury [26]. Normally, circulating fatty acids are bound to albumin, where they are much less toxic than free fatty acids. If microvascular obstruction from the embolization
procedure limits the available pool of albumin locally, this could result in higher concentrations of free fatty acids, resulting in lung injury [26]. This hypothesis is also supported by our finding that hypoalbuminemia is a risk factor for TAE/TACE-induced POE. It is also interesting to speculate that the higher grade of liver disease as a risk factor for TAE/TACE-induced POE may be a reflection of lower serum albumin levels. In our study, patients with POE had a high mortality rate. Five patients died as a result of multiorgan failure. Hsu et al. [27] found that in HCC patients with ascites, hypoalbuminemia was associated with the occurrence of post-TACE acute renal failure, a poor prognostic predictor in this subset of HCC patients. In the present cohort, patients with higher grade cirrhosis who developed POE had a greater frequency of acute renal failure. Worsening hepatic function may also be observed after TACE. Risk factors for TACE-induced hepatic failure include elevated serum bilirubin levels, portal vein thrombosis, poor baseline hepatic function, high-dose infusion of chemotherapeutic agents, and a history of multiple embolization procedures [28,29]. Our study has several limitations. This study was retrospective and included a relatively small numbers of patients in the POE group. However, our criteria for the clinical diagnosis of POE were strict and we likely excluded some patients who probably had POE. For example, patients with comorbidities – such as congestive heart failure, respiratory infections, and pre-existing chronic lung disorders – that may mimic POE in its presentation were excluded. Indeed, Clouse et al. [30] reported the presence of POE in the majority of patients who had undergone lymphography. In patients with respiratory symptoms after TAE or TACE without obvious infiltration on chest radiography, we also did not consistently obtain CT to exclude POE. CT scanning is likely significantly more sensitive in diagnosing POE, as evidenced by a CT surveillance study that showed that lipiodol accumulated in the lung parenchyma and pleura in 81% of their patients by 24 h after administration of TACE. This finding suggests that lipid embolization occurs early and frequently after TACE but only a subset becomes clinically evident [31]. The actual incidence of POE is not known, as mild cases often go unnoticed. We used stricter criteria for diagnosing POE, but the incidence was almost 10%. We think POE may be quite common but only a subset of patients develop clinically evident POE. This is similar to the fat emboli syndrome after long bone fracture. The incidence of fat embolism syndrome ranges from less than 1 to 29% in different studies; however, only a small subset of patients are symptomatic, resulting in the fat embolism syndrome [32]. Conclusion
Lipiodol dosage used in TAE or TACE to treat hepatic malignancies appears to be a robust predictor of the development of clinically significant POE. Although our
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Risk factors for development of POE Wu et al.
findings suggest keeping the lipiodol dose to 14.5 ml or lower to reduce the possibility of POE, a larger, prospective study is required to confirm this.
14
Acknowledgements
16
Conflicts of interest
17
15
There are no conflicts of interest. 18
References 1
2
3 4
5
6
7
8
9
10
11
12
13
Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM. Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int J Cancer 2010; 127:2893–2917. Llovet JM, Real MI, Montan˜a X, Planas R, Coll S, Aponte J, et al. Arterial embolisation or chemoembolisation versus symptomatic treatment in patients with unresectable hepatocellular carcinoma: a randomised controlled trial. Lancet 2002; 359:1734–1739. Forner A, Llovet JM, Bruix J. Hepatocellular carcinoma. Lancet 2012; 379:1245–1255. Oliveri RS, Wetterslev J, Gluud C. Transarterial (chemo)embolisation for unresectable hepatocellular carcinoma. Cochrane Database Syst Rev 2011; 3:CD004787. Ono Y, Yoshimasu T, Ashikaga R, Inoue M, Shindou H, Fuji K, et al. Long-term results of lipiodol-transcatheter arterial embolization with cisplatin or doxorubicin for unresectable hepatocellular carcinoma. Am J Clin Oncol 2000; 23:564–568. Ueno K, Miyazono N, Inoue H, Nishida H, Kanetsuki I, Nakajo M. Transcatheter arterial chemoembolization therapy using iodized oil for patients with unresectable hepatocellular carcinoma: evaluation of three kinds of regimens and analysis of prognostic factors. Cancer 2000; 88:1574–1581. Sakamoto I, Aso N, Nagaoki K, Matsuoka Y, Uetani M, Ashizawa K, et al. Complications associated with transcatheter arterial embolization for hepatic tumors. Radiographics 1998; 18:605–619. Bismuth H, Morino M, Sherlock D, Castaing D, Miglietta C, Cauquil P, et al. Primary treatment of hepatocellular carcinoma by arterial chemoembolization. Am J Surg 1992; 163:387–394. Venook AP, Stagg RJ, Lewis BJ, Chase JL, Ring EJ, Maroney TP, et al. Chemoembolization for hepatocellular carcinoma. J Clin Oncol 1990; 8:1108–1114. Stuart K, Stokes K, Jenkins R, Trey C, Clouse M. Treatment of hepatocellular carcinoma using doxorubicin/ethiodized oil/gelatin powder chemoembolization. Cancer 1993; 72:3202–3209. Chung JW, Park JH, Im JG, Han JK, Han MC. Pulmonary oil embolism after transcatheter oily chemoembolization of hepatocellular carcinoma. Radiology 1993; 187:689–693. Yamaura K, Higashi M, Akiyoshi K, Itonaga Y, Inoue H, Takahashi S. Pulmonary lipiodol embolism during transcatheter arterial chemoembolization for hepatoblastoma under general anaesthesia. Eur J Anaesthesiol 2000; 17:704–708. Czauderna P, Zbrzezniak G, Narozanski W, Sznurkowska K, SkoczylasStoba B, Stoba C. Pulmonary embolism: a fatal complication of arterial chemoembolization for advanced hepatocellular carcinoma. J Pediatr Surg 2005; 40:1647–1650.
19
20
21
22
23
24
25
26
27
28
29
30
31
32
981
Wu GC, Perng WC, Chen CW, Chian CF, Peng CK, Su WL. Acute respiratory distress syndrome after transcatheter arterial chemoembolization of hepatocellular carcinomas. Am J Med Sci 2009; 338:357–360. Bruix J, Sherman M. Management of hepatocellular carcinoma. Hepatology 2005; 42:1208–1236. Bruix J, Sherman M. Management of hepatocellular carcinoma: an update. Hepatology 2011; 53:1020–1022. Luo MY, Shan H, Jiang ZB, Li LF, Huang HQ. Study on hepatocellular carcinoma-associated hepatic arteriovenous shunt using multidetector CT. World J Gastroenterol 2003; 9:2455–2459. Xia J, Ren Z, Ye S, Sharma D, Lin Z, Gan Y, et al. Study of severe and rare complications of transarterial chemoembolization (TACE) for liver cancer. Eur J Radiol 2006; 59:407–412. Walser RH, Haldemann AR, Ro¨sler H, Koranda P, Kinser JA, Triller J. Diagnostic angioscintigraphic evaluation of malignant hepatic tumors before catheter embolization: determination of shunt, flow distribution, and reflux. Cardiovasc Intervent Radiol 1996; 19:77–81. Leung WT, Lau WY, Ho SK, Chan M, Leung NW, Lin J, et al. Measuring lung shunting in hepatocellular carcinoma with intrahepatic-arterial technetium99m macroaggregated albumin. J Nucl Med 1994; 35:70–73. Ho S, Lau WY, Leung WT, Chan M, Chan KW, Johnson PJ, et al. Arteriovenous shunts in patients with hepatic tumors. J Nucl Med 1997; 38:1201–1205. Kudo S, Wright KC, Chuang VP, Wallace S, Mir S, Bechtel W. Experimental evaluation of intraarterial occlusion-infusion chemotherapy. AJR Am J Roentgenol 1984; 143:1069–1073. Kishi K, Sonomura T, Satoh M, Nishida N, Terada M, Shioyama Y, et al. Acute toxicity of lipiodol infusion into the hepatic arteries of dogs. Invest Radiol 1994; 29:882–889. Wu JJ, Chao M, Zhang GQ, Li B, Dong F. Pulmonary and cerebral lipiodol embolism after transcatheter arterial chemoembolization [corrected] in hepatocellular carcinoma. World J Gastroenterol 2009; 15:633–635. Zhao H, Wang HQ, Fan QQ, Chen XX, Lou JY. Rare pulmonary and cerebral complications after transarterial chemoembolisation for hepatocellular carcinoma: a case report. World J Gastroenterol 2008; 14:6425–6427. Silvestri RC, Huseby JS, Rughani I, Thorning D, Culver BH. Respiratory distress syndrome from lymphangiography contrast medium. Am Rev Respir Dis 1980; 122:543–549. Hsu CY, Huang YH, Su CW, Lin HC, Chiang JH, Lee PC, et al. Renal failure in patients with hepatocellular carcinoma and ascites undergoing transarterial chemoembolization. Liver Int 2010; 30:77–84. Jeon SH, Park KS, Kim YH, Shin YS, Kang MK, Jang BK, et al. Incidence and risk factors of acute hepatic failure after transcatheter arterial chemoembolization for hepatocellular carcinoma. Korean J Gastroenterol 2007; 50:176–182. Katsushima S, Inokuma T, Oi H, Okamura J, Higashi T, Takeuchi R, et al. Acute hepatic failure following transcatheter arterial embolization for the treatment of hepatocellular carcinoma. Digestion 1997; 58:189–195. Clouse ME, Hallgrimsson J, Wenlund DE. Complications following lymphography with particular reference to pulmonary oil embolization. Am J Roentgenol Radium Ther Nucl Med 1966; 96:972–978. Lo¨rcher U, Peters J, Kollath J. Changes in the lungs and pleura following chemoembolization of liver tumors with mitomycin-lipiodol. Rofo 1990; 152:569–73. Shaikh N. Emergency management of fat embolism syndrome. J Emerg Trauma Shock 2009; 2:29–33.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Risk factors for the development of pulmonary oil embolism after transcatheter arterial chemoembolization of hepatic tumors Geng-Chin Wua, Edward D. Chanh,i,j, Yu-Ching Choub, Chih-Yung Yuc, Tsai-Yuan Hsiehd, Chung-Bao Hsiehe, Chih-Feng Chianf, Fu-Chang Kea, Yu-Ling Daif and Wen-Lin Sub,f,g Pulmonary oil embolism (POE) is a rare fatal complication after transcatheter arterial embolization (TAE) and transcatheter arterial chemoembolization (TACE). As risk factors have not been clearly delineated, the aim of the present study was to identify the risk factors for development of POE after TACE. A retrospective analysis was carried out on patients with unresectable hepatocellular carcinoma who received TAE or TACE at the Tri-Service General Hospital (Taiwan) between January 2005 and December 2008. The diagnosis of TAE-induced or TACE-induced POE was based on development of respiratory signs and symptoms relatively soon after the procedure, as well as based on characteristic radiographic findings. Of the 219 enrolled patients in this study, 20 were diagnosed with POE after TAE or TACE. On univariate logistic regression analysis, patients developing POE were found to be older (67.95±15.95 vs. 61.44±12.59 years, P = 0.033), with a lower serum albumin level (3.25±0.58 vs. 3.62±0.57 g/dl, P = 0.009), a higher grade of liver cirrhosis as classified on the basis of Child’s criteria (P < 0.006), a larger tumor size (8.55±4.52 vs. 4.78±3.97 cm in diameter, P < 0.001), a higher lipioidol dose (22.35±11.01 vs. 13.69±7.66 ml, P = 0.003), and a higher doxorubicin dose (50.27±7.05 vs. 40.75±13.61 mg, P < 0.001). Following multivariate logistic regression analysis, only lipiodol dose was found to be a significant risk factor for POE (odds ratio = 1.133, 95% confidence interval: 1.004, 1.279; P = 0.044). The receiver operator characteristic curve cutoff
Introduction Hepatocellular carcinoma (HCC) is one of the most common malignant tumors in man and one of the deadliest [1]. Transcatheter arterial embolization (TAE) and transcatheter arterial chemoembolization (TACE) are relatively effective palliative treatments for unresectable HCC. Although TACE has been shown to increase survival [2,3], a recent meta-analysis found that the benefit of TACE was still limited [4]. Ethiodized oil (lipiodol), comprising iodine combined with ethyl esters of fatty acids of poppyseed oil, has been widely used as an embolic agent and a carrier of chemotherapeutic agents in the treatment of HCC [5,6]. Various complications arising during TACE have been reported, including acute hepatic failure, liver abscess, intrahepatic biloma formation, liver c 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins 0959-4973
point for lipiodol dose level was 14.5 ml, with a sensitivity of 80% and a specificity of 66.3%. In conclusion, the lipiodol dose could be considered as a predictive factor for POE after TAE or TACE in hepatic malignant tumor patients. On the basis of this retrospective study, the safe lipiodol dose to minimize the risk for POE is 14.5 ml or lower; however, larger, prospective studies are needed to determine the optimally safe and yet efficacious c 2014 Wolters dose. Anti-Cancer Drugs 25:976–981 Kluwer Health | Lippincott Williams & Wilkins. Anti-Cancer Drugs 2014, 25:976–981 Keywords: hepatocellular carcinoma, pulmonary oil embolism, transcatheter arterial chemoembolization a Department of Internal Medicine, Division of Pulmonary Medicine, Taoyuan Armed Forces General Hospital, Lungtan, Taoyuan, bSchool of Public Health , National Defense Medical Center, cDepartment of Radiology, dDepartment of Internal Medicine, Division of Gastroenterology, eDepartment of Surgery, Division of Organ Transplantation, fDepartment of Medicine, Division of Pulmonary and Critical Care Medicine, Tri-Service General Hospital, Nei-Hu, Taipei, gDepartment of Medicine, Division of Pulmonary and Critical Care Medicine, Taipei Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation, New Taipei, Taiwan, Republic of China, hDivision of Pulmonary Sciences and Critical Care Medicine, Department of Medicine, University of Colorado Denver at Anschutz Medical Center, iDenver Veterans Affairs Medical Center and jDepartment of Medicine, Division of the Mycobacterial and Respiratory Infections, National Jewish Health, Denver, Colorado, USA
Correspondence to Wen-Lin Su, MD, PhD, Department of Medicine, Division of Pulmonary and Critical Care Medicine, Taipei Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation, No.289, Jianguo Rd, Xindian Dist., New Taipei City 231, Taiwan, Republic of China Tel: + 886 2 8792 7233; fax: + 886 2 6600 8692; e-mail: [email protected] Received 2 November 2013 Revised form accepted 8 March 2014
infarction, multiple intrahepatic aneurysms, gallbladder infarction, cholecystitis, splenic infarction, gastrointestinal mucosal lesions, pulmonary oil embolism (POE), tumor rupture, and variceal bleeding [7–10]. POE is an extremely rare but serious complication after TACE [7,11]. Sakamoto et al. [7] reported that POE occurred with a frequency of 0.17% after TACE. The amount of lipiodol infused and the presence of arteriovenous shunting in the tumor have been shown to increase the risk for POE [11–13]. Chung et al. [11] recommended that the maximum safe dose of lipiodol is 15–20 ml/patient (B0.25 ml/kg). Nevertheless, patients without arteriovenous shunts developing POE while receiving recommended doses of lipiodol for TACE have been reported [14]. DOI: 10.1097/CAD.0000000000000113
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Risk factors for development of POE Wu et al. 977
Identifying additional risk factors for the development of POE after TAE/TACE can be clinically useful for, perhaps, instituting preventive measures such as dose reduction in high risk individuals. The present study aimed to evaluate the clinical feature of patients with POE and to identify the risk factors for its development.
Materials and methods Patients
In this single-center, retrospective study, consecutive patients who underwent TAE or TACE between January 2005 and December 2008 at Tri-Service General Hospital for unresectable HCC or metastatic cancer of the liver were analyzed for the development of POE. The study was approved by the Internal Review Board at Tri-Service General Hospital. The diagnosis of HCC was confirmed through histologic and cytologic findings or abdominal multidetector computed tomography (MDCT) findings consistent with HCC [15,16]. Cases of metastatic cancer to the liver were confirmed histopathologically.
by the development respiratory signs and symptoms (cough, dyspnea, hemoptysis, and tachypnea), reduced gas exchange (decreased oxygen saturation on arterial blood gas measurements), and the presence of ill-defined, high-density reticular opacities on chest radiographs (Fig. 1) and/or chest computed tomography (CT) scans (Fig. 2) within 7 days after administration of TAE or TACE. As the clinical manifestations of POE are nonspecific and may mimic other diseases such as heart failure, infectious pneumonia, chronic airway or interstitial lung disease, and transfusion-related acute lung injury, we also stipulated that patients classified as having POE have normal left ventricle ejection fraction on echocardiography, negative sputum and blood cultures, no clinical evidence of fluid overload, no history of acute or chronic lung diseases such as asthma, chronic obstructive pulmonary disease, and interstitial lung disease, no recent transfusion of any blood products (before developing respiratory symptoms after TAE or TACE), and no infiltration on chest radiographs before TAE or TACE. Statistical analysis
Chemoembolization
All participating radiologists had prior experience with lipiodol chemoembolization. Hepatic angiography and MDCT were routinely performed to delineate the blood supply, the arteriovenous shunt, and the size and extent of the HCC. TAE was initiated with the infusion of iodized oil (lipiodol; Guerbet, Aulnay-sous-Bois, France) through the proper hepatic artery, left hepatic artery, or right hepatic artery, which supplied the tumors. For TACE, an emulsion consisting of lipiodol and doxorubicin hydrochloride (adriamycin; Pfizer Italia S.R.L., Milano, Italy) was infused. The amount of iodized oil and doxorubicin hydrochloride used ranged from 4 to 50 ml and 6 to 60 mg, respectively. These dosages were determined on the basis of tumor size and vascularity.
Patients were categorized into groups with and without POE following TAE or TACE. As most patients had undergone TAE or TACE more than once, the data were obtained following the first administration of TAE or TACE. Continuous and categorical variables are presented as mean±SD and number (percentage), respectively. Fig. 1
Embolization was subsequently performed with gelatinsponge particles (Spongostan standard; Johnson & Johnson Medical Pty. Ltd, Gauteng, South Africa) until flow in the hepatic artery was sluggish. The goal was to decrease arterial flow without causing total obstruction. Data collection
The following data were extracted from patient medical records: demographic characteristics (including age and sex), body weight, results from standard blood analysis (including white blood cell count and C-reactive protein, aspartate aminotransferase, alanine aminotransferase, total bilirubin, albumin, alkaline phosphatase, and g-glutamyl transpeptidase levels), etiology of hepatic tumor, Child’s classification of the liver before treatment, tumor size, method of embolization (TAE or TACE), and dosages of lipiodol (ml) and doxorubicin (mg). Definition and diagnosis
Chest radiographs obtained during admission were analyzed by a chest radiologist and a pulmonologist. POE was defined
Chest radiograph showing high-density ill-defined reticular opacities of both lungs.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
978 Anti-Cancer Drugs 2014, Vol 25 No 8
Fig. 2
Fig. 3
ROC curve 1.0
0.8
Sensitivity
0.6
0.4
0.2
0.0 0.0
0.2
0.4
0.6
0.8
1.0
1−Specificity Noncontrast CT scan showing linear high-density shadows (black arrow) in both lung fields, indicating lipiodol retention. CT, computed tomography.
P-values were determined by t-test or w2/Fisher’s Exact Test, by Fisher’s Exact Test, or by w2-test. Univariate logistic regression analysis and multivariate logistic regression analysis were carried out on continuous variables to identify factors of POE associated with TAE or TACE. A receiver operator characteristic curve (Fig. 3) was plotted to determine the maximum safe dose of lipiodol. Analyses were carried out using the SPSS 17.0 software package (SPSS Institute Inc., Chicago, Illinois, USA). All statistical assessments were two-sided, using a significance level of 0.05.
Receiver operator characteristic (ROC) curve for lipiodol dose level.
Fig. 4
Patients with unresectable hepatocellular carcinoma who underwent TAE or TACE (219 elevated)
Patients had respiratory symptoms after TAE or TACE (n = 25)
Results Of the 219 patients who underwent TAE or TACE, 205 had HCC and 14 had metastatic cancer to the liver. Respiratory signs and symptoms, including cough, hemoptysis, dyspnea, and/or tachypnea, developed within 1 h to 6 days after TACE and TAE. Twenty (9.1%) of the 219 patients met the criteria of having POE after the procedure (Fig. 4). Whereas five additional patients did have respiratory symptoms after TAE or TACE, they did not show high-density reticular opacities in the lungs on imaging and thus were not defined as having POE. All 219 patients underwent angiography and MDCT. Interestingly, these imaging studies did not reveal evidence of arteriovenous shunting in any of the patients. Table 1 lists the demographic characteristics and laboratory values of the 219 patients. The cause of HCC was almost always chronic hepatitis B or chronic
No infiltration on chest X-ray (n = 5)
Patients with pulmonary oil embolism after TAE or TACE (n = 20)
Patients without pulmonary oil embolism after TAE or TACE (n = 199)
Patients who received TAE or TACE and who had clinical symptoms and radiographic abnormalities consistent with pulmonary oil embolism. TACE, transcatheter arterial chemoembolization; TAE, transcatheter arterial embolization.
hepatitis C infections in those with or without POE. Patients who developed POE after TAE or TACE procedures were found to be older (67.95±15.95 vs.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Risk factors for development of POE Wu et al. 979
Comparison of demographics and clinical features between pulmonary embolism and non-pulmonary embolisma
Table 2 Univariate logistic regression analysis of pulmonary embolism rate
Table 1
Variables
PE (n = 20) [N (%)]
Non-PE (n = 199) [N (%)]
P-valueb d
Sex 1.000 Male 13 (65.0) 128 (64.3) Female 7 (35.0) 71 (35.7) Age 67.95±15.95 61.44±12.59 0.033 Body weight (kg) 61.6±14.2 64.45±10.81 0.277 Laboratory data WBC (/ml) 6719.0±3296.0 5298.92±2170.38 0.074 CRP (mg/dl) 15.27±7.90 11.52±10.58 0.363 AST (U/l) 105.55±100.85 76.85±75.18 0.117 ALT (U/l) 71.8±68.5 67.15±63.26 0.756 T.bil (mg/dl) 1.56±1.86 1.18±1.27 0.223 Alb (g/dl) 3.25±0.58 3.62±0.57 0.009 ALP (U/l) 272.77±299.10 124.19±92.56 0.100 g-GT (U/l) 198.7±158.9 137.71±136.56 0.265 Etiology 0.960c Alcohol 0 (0) 6 (3.0) HBV 11 (55.0) 108 (54.3) HCV 8 (40.0) 67 (33.7) META 1 (5.0) 13 (6.5) Unknown 0 5(2.5) Child’s 0.006c classification of liver cirrhosis A 10 (50.0) 161 (80.9) B 9 (45.0) 36 (18.1) C 0 1 (0.5) N 1 (5.0) 1 (0.5) Tumor size (cm) 8.55±4.52 4.78±3.97 < 0.001 Lipiodol dose 22.35±11.01 13.69±7.66 0.003 (ml) Adriamycin (mg) 50.27±7.05 (n = 15) 40.75±13.61 (n = 148) < 0.001
Alb, albumin; ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CRP, C-reactive protein; HBV, hepatitis B virus; HCV, hepatitis C virus; META, metastatic; N, non-cirrhosis of liver; T.bil, total bilirubin; WBC, white blood cell; g-GT, g-glutamyl transpeptidase. a Continuous and categorical variables are presented as mean±SD and number (%), respectively. b P-value by t-test or w2/Fisher’s exact test. c Fisher’s exact test. d 2 w -test.
61.44±12.59 years, P = 0.033), with lower serum albumin levels (3.25±0.58 vs. 3.62±0.57 g/dl, P < 0.01), more advanced stage of cirrhosis based on Child’s classification (P < 0.006), and larger tumor size (8.55±4.52 vs. 4.78± 3.97 cm in diameter, P < 0.001). Patients who developed POE also were found to have received higher doses of lipiodol (22.35±11.01 vs. 13.69±7.66 ml, P = 0.003) and doxorubicin (50.27±7.05 vs. 40.75±13.61 mg, P < 0.001).
Variables Sex (M/F) Age Body weight (kg) Laboratory data WBC (/ml) CRP (mg/dl) AST (U/l) ALT (U/l) T.bil (mg/dl) Alb (g/dl) ALP (U/l) g-GT (U/l) Etiology Others (alcohol + META + Unknown) HBV HCV Non-HBV (HCV + alcohol + META + unknown) HBV Child’s classification of liver cirrhosis A+N B+C Tumor size (cm) Lipiodol dose (ml) Adriamycin (mg)
Odds ratio
95% Confidence interval
1.030 1.047 0.976
0.393, 2.700 1.003, 1.092 0.935, 1.019
0.952 0.034 0.277
1.002 1.038 1.003 1.001 1.159 0.328 1.005 1.002
1.0015, 1.0024 0.960, 1.122 0.999, 1.007 0.994, 1.008 0.905, 1.486 0.139, 0.776 1.001, 1.008 0.998, 1.007
0.012 0.355 0.140 0.755 0.242 0.011 0.009 0.274
1.0
P-value
Reference
2.444 2.866 1.0
0.301, 19.849 0.340, 24.126 Reference
0.403 0.333
1.030
0.409, 2.595
0.950
1.0 3.582 1.167 1.098 1.070
Reference 1.385, 9.268 1.07, 1.273 1.048, 1.151 1.014, 1.13
0.009 < 0.001 < 0.001 0.013
Alb, albumin; ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CRP, C-reactive protein; HBV, hepatitis B virus; HCV, hepatitis C virus; META, metastatic; N, non-cirrhosis of liver; T.bil, total bilirubin; WBC, white blood cell; g-GT, g-glutamyl transpeptidase. P < 0.05.
Table 3 Multivariate logistic regression analysis of pulmonary embolisma Variables
Odds ratio
Age 1.039 Laboratory data WBC (/ml) 1.0001 Alb (g/dl) 2.160 ALP (U/l) 1.003 Child’s classification of liver cirrhosis A+N 1.0 B+C 2.462 Tumor size (cm) 1.028 Lipiodol dose (ml) 1.133 Adriamycin (mg) 1.031
95% Confidence interval
P-value
0.964, 1.119
0.316
0.9998, 1.0004 0.334, 13.979 0.998, 1.007
0.613 0.419 0.222
Reference 0.247, 24.552 0.862, 1.226 1.004, 1.279 0.950, 1.118
0.442 0.759 0.044 0.470
Alb, albumin; ALP, alkaline phosphatase; WBC, white blood cell. The statistically significantly factors (P < 0.05), including the continuous and categorical variables from Table 2. a
Table 2 shows the results of univariate logistic regression analysis. Increasing age, white blood cell count, alkaline phosphatase level, tumor size, lipiodol dose, and doxorubicin dose, as well as severity of cirrhosis, were associated with increased risk for POE. In contrast, increased serum albumin levels were associated with lower risk for POE. Following multivariate logistic regression analysis, only the dose of lipiodol was found to be an independent risk factor for the development of POE (Table 3). On the basis of a receiver operator characteristic curve for lipiodol dose and POE, lipiodol dose higher than 14.5 ml was found to diagnose POE with a sensitivity of 80% and a
specificity of 66.3%. The overall mortality rate for patients with POE was 25%; the five patients who died had developed multiorgan failure, including respiratory failure, hepatic failure, and oliguria with acute renal failure.
Discussion In this retrospective analysis of 219 patients who underwent TAE or TACE for unresectable malignant tumor of the liver, we found, on multivariate regression analysis, that the lipiodol dose was the primary risk factor for POE. It makes intuitive sense that the amount of lipiodol
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
980 Anti-Cancer Drugs 2014, Vol 25 No 8
infused is a risk factor for the development of POE, as has been suggested by others [11–13]. Indeed, Chung et al. [11] recommended that the maximum safe dose of lipiodol is 15–20 ml/patient (B0.25 ml/kg). Our finding that a lipiodol dose greater than 14.5 ml is associated with the development of POE would lend credence to this recommendation. Intratumor arteriovenous shunting is considered to be a risk factor for POE. MDCT is an important medical imaging examination technology to detect the hepatic arteriovenous shunt associated with HCC [17]. In the present study, all patients underwent both angiography and MDCT, and arteriovenous shunts were not detected in any patient, a finding consistent with those of previous studies [11–14,18]. The inability to detect arteriovenous shunts may be related to the decreased sensitivity of angiography to detect microvascular shunts and thus the possibility that the presence of such shunts predispose to POE is not ruled out. The technetium-99m-macroaggregated albumin test may be more sensitive in detecting lung shunting in HCC [19,20]. Ho et al. [21] demonstrated that lung shunting was correlated with tumor size and tumor vascularity in patients with HCC. Although our univariate regression analysis did show an increased risk for POE based on tumor size, multivariate analysis failed to show tumor size as an independent risk factor for POE. Intra-arterial infusion chemotherapy with doxorubicin is reported to be complicated by thrombosis of regional blood vessels and local toxicity [22]. Although the results of univariate logistic regression analysis in our study show that the dose of doxorubicin is a risk factor for POE, multivariate analysis failed to demonstrate this. The route by which oil droplets injected into the liver accumulate in the pulmonary vasculature is not precisely known. Lipiodol infusion into the hepatic artery of dogs resulted in dose-dependent POE. These observations substantiated embolization of oil droplets through the hepatic sinusoids to pulmonary capillaries [23]. Further, POE was probably associated with hepatic vein invasion or communication between the tumor feeding artery and the pulmonary vein that might occur as a result of tumor invasion into the pleura or diaphragm [24,25]. Thus, it is reasonable to posit that the pathogenesis of POE likely involves migration of oil droplets from the hepatic sinusoids into the pulmonary capillaries through microvascular shunts, hepatic vein invasion, or tumor invasion into the diaphragm or pleural space. The mechanism by which lipiodol itself causes pulmonary toxicity is not well understood, although direct chemical injury has been postulated, as free fatty acids injected intravenously result in the rapid onset of severe lung injury [26]. Normally, circulating fatty acids are bound to albumin, where they are much less toxic than free fatty acids. If microvascular obstruction from the embolization
procedure limits the available pool of albumin locally, this could result in higher concentrations of free fatty acids, resulting in lung injury [26]. This hypothesis is also supported by our finding that hypoalbuminemia is a risk factor for TAE/TACE-induced POE. It is also interesting to speculate that the higher grade of liver disease as a risk factor for TAE/TACE-induced POE may be a reflection of lower serum albumin levels. In our study, patients with POE had a high mortality rate. Five patients died as a result of multiorgan failure. Hsu et al. [27] found that in HCC patients with ascites, hypoalbuminemia was associated with the occurrence of post-TACE acute renal failure, a poor prognostic predictor in this subset of HCC patients. In the present cohort, patients with higher grade cirrhosis who developed POE had a greater frequency of acute renal failure. Worsening hepatic function may also be observed after TACE. Risk factors for TACE-induced hepatic failure include elevated serum bilirubin levels, portal vein thrombosis, poor baseline hepatic function, high-dose infusion of chemotherapeutic agents, and a history of multiple embolization procedures [28,29]. Our study has several limitations. This study was retrospective and included a relatively small numbers of patients in the POE group. However, our criteria for the clinical diagnosis of POE were strict and we likely excluded some patients who probably had POE. For example, patients with comorbidities – such as congestive heart failure, respiratory infections, and pre-existing chronic lung disorders – that may mimic POE in its presentation were excluded. Indeed, Clouse et al. [30] reported the presence of POE in the majority of patients who had undergone lymphography. In patients with respiratory symptoms after TAE or TACE without obvious infiltration on chest radiography, we also did not consistently obtain CT to exclude POE. CT scanning is likely significantly more sensitive in diagnosing POE, as evidenced by a CT surveillance study that showed that lipiodol accumulated in the lung parenchyma and pleura in 81% of their patients by 24 h after administration of TACE. This finding suggests that lipid embolization occurs early and frequently after TACE but only a subset becomes clinically evident [31]. The actual incidence of POE is not known, as mild cases often go unnoticed. We used stricter criteria for diagnosing POE, but the incidence was almost 10%. We think POE may be quite common but only a subset of patients develop clinically evident POE. This is similar to the fat emboli syndrome after long bone fracture. The incidence of fat embolism syndrome ranges from less than 1 to 29% in different studies; however, only a small subset of patients are symptomatic, resulting in the fat embolism syndrome [32]. Conclusion
Lipiodol dosage used in TAE or TACE to treat hepatic malignancies appears to be a robust predictor of the development of clinically significant POE. Although our
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Risk factors for development of POE Wu et al.
findings suggest keeping the lipiodol dose to 14.5 ml or lower to reduce the possibility of POE, a larger, prospective study is required to confirm this.
14
Acknowledgements
16
Conflicts of interest
17
15
There are no conflicts of interest. 18
References 1
2
3 4
5
6
7
8
9
10
11
12
13
Ferlay J, Shin HR, Bray F, Forman D, Mathers C, Parkin DM. Estimates of worldwide burden of cancer in 2008: GLOBOCAN 2008. Int J Cancer 2010; 127:2893–2917. Llovet JM, Real MI, Montan˜a X, Planas R, Coll S, Aponte J, et al. Arterial embolisation or chemoembolisation versus symptomatic treatment in patients with unresectable hepatocellular carcinoma: a randomised controlled trial. Lancet 2002; 359:1734–1739. Forner A, Llovet JM, Bruix J. Hepatocellular carcinoma. Lancet 2012; 379:1245–1255. Oliveri RS, Wetterslev J, Gluud C. Transarterial (chemo)embolisation for unresectable hepatocellular carcinoma. Cochrane Database Syst Rev 2011; 3:CD004787. Ono Y, Yoshimasu T, Ashikaga R, Inoue M, Shindou H, Fuji K, et al. Long-term results of lipiodol-transcatheter arterial embolization with cisplatin or doxorubicin for unresectable hepatocellular carcinoma. Am J Clin Oncol 2000; 23:564–568. Ueno K, Miyazono N, Inoue H, Nishida H, Kanetsuki I, Nakajo M. Transcatheter arterial chemoembolization therapy using iodized oil for patients with unresectable hepatocellular carcinoma: evaluation of three kinds of regimens and analysis of prognostic factors. Cancer 2000; 88:1574–1581. Sakamoto I, Aso N, Nagaoki K, Matsuoka Y, Uetani M, Ashizawa K, et al. Complications associated with transcatheter arterial embolization for hepatic tumors. Radiographics 1998; 18:605–619. Bismuth H, Morino M, Sherlock D, Castaing D, Miglietta C, Cauquil P, et al. Primary treatment of hepatocellular carcinoma by arterial chemoembolization. Am J Surg 1992; 163:387–394. Venook AP, Stagg RJ, Lewis BJ, Chase JL, Ring EJ, Maroney TP, et al. Chemoembolization for hepatocellular carcinoma. J Clin Oncol 1990; 8:1108–1114. Stuart K, Stokes K, Jenkins R, Trey C, Clouse M. Treatment of hepatocellular carcinoma using doxorubicin/ethiodized oil/gelatin powder chemoembolization. Cancer 1993; 72:3202–3209. Chung JW, Park JH, Im JG, Han JK, Han MC. Pulmonary oil embolism after transcatheter oily chemoembolization of hepatocellular carcinoma. Radiology 1993; 187:689–693. Yamaura K, Higashi M, Akiyoshi K, Itonaga Y, Inoue H, Takahashi S. Pulmonary lipiodol embolism during transcatheter arterial chemoembolization for hepatoblastoma under general anaesthesia. Eur J Anaesthesiol 2000; 17:704–708. Czauderna P, Zbrzezniak G, Narozanski W, Sznurkowska K, SkoczylasStoba B, Stoba C. Pulmonary embolism: a fatal complication of arterial chemoembolization for advanced hepatocellular carcinoma. J Pediatr Surg 2005; 40:1647–1650.
19
20
21
22
23
24
25
26
27
28
29
30
31
32
981
Wu GC, Perng WC, Chen CW, Chian CF, Peng CK, Su WL. Acute respiratory distress syndrome after transcatheter arterial chemoembolization of hepatocellular carcinomas. Am J Med Sci 2009; 338:357–360. Bruix J, Sherman M. Management of hepatocellular carcinoma. Hepatology 2005; 42:1208–1236. Bruix J, Sherman M. Management of hepatocellular carcinoma: an update. Hepatology 2011; 53:1020–1022. Luo MY, Shan H, Jiang ZB, Li LF, Huang HQ. Study on hepatocellular carcinoma-associated hepatic arteriovenous shunt using multidetector CT. World J Gastroenterol 2003; 9:2455–2459. Xia J, Ren Z, Ye S, Sharma D, Lin Z, Gan Y, et al. Study of severe and rare complications of transarterial chemoembolization (TACE) for liver cancer. Eur J Radiol 2006; 59:407–412. Walser RH, Haldemann AR, Ro¨sler H, Koranda P, Kinser JA, Triller J. Diagnostic angioscintigraphic evaluation of malignant hepatic tumors before catheter embolization: determination of shunt, flow distribution, and reflux. Cardiovasc Intervent Radiol 1996; 19:77–81. Leung WT, Lau WY, Ho SK, Chan M, Leung NW, Lin J, et al. Measuring lung shunting in hepatocellular carcinoma with intrahepatic-arterial technetium99m macroaggregated albumin. J Nucl Med 1994; 35:70–73. Ho S, Lau WY, Leung WT, Chan M, Chan KW, Johnson PJ, et al. Arteriovenous shunts in patients with hepatic tumors. J Nucl Med 1997; 38:1201–1205. Kudo S, Wright KC, Chuang VP, Wallace S, Mir S, Bechtel W. Experimental evaluation of intraarterial occlusion-infusion chemotherapy. AJR Am J Roentgenol 1984; 143:1069–1073. Kishi K, Sonomura T, Satoh M, Nishida N, Terada M, Shioyama Y, et al. Acute toxicity of lipiodol infusion into the hepatic arteries of dogs. Invest Radiol 1994; 29:882–889. Wu JJ, Chao M, Zhang GQ, Li B, Dong F. Pulmonary and cerebral lipiodol embolism after transcatheter arterial chemoembolization [corrected] in hepatocellular carcinoma. World J Gastroenterol 2009; 15:633–635. Zhao H, Wang HQ, Fan QQ, Chen XX, Lou JY. Rare pulmonary and cerebral complications after transarterial chemoembolisation for hepatocellular carcinoma: a case report. World J Gastroenterol 2008; 14:6425–6427. Silvestri RC, Huseby JS, Rughani I, Thorning D, Culver BH. Respiratory distress syndrome from lymphangiography contrast medium. Am Rev Respir Dis 1980; 122:543–549. Hsu CY, Huang YH, Su CW, Lin HC, Chiang JH, Lee PC, et al. Renal failure in patients with hepatocellular carcinoma and ascites undergoing transarterial chemoembolization. Liver Int 2010; 30:77–84. Jeon SH, Park KS, Kim YH, Shin YS, Kang MK, Jang BK, et al. Incidence and risk factors of acute hepatic failure after transcatheter arterial chemoembolization for hepatocellular carcinoma. Korean J Gastroenterol 2007; 50:176–182. Katsushima S, Inokuma T, Oi H, Okamura J, Higashi T, Takeuchi R, et al. Acute hepatic failure following transcatheter arterial embolization for the treatment of hepatocellular carcinoma. Digestion 1997; 58:189–195. Clouse ME, Hallgrimsson J, Wenlund DE. Complications following lymphography with particular reference to pulmonary oil embolization. Am J Roentgenol Radium Ther Nucl Med 1966; 96:972–978. Lo¨rcher U, Peters J, Kollath J. Changes in the lungs and pleura following chemoembolization of liver tumors with mitomycin-lipiodol. Rofo 1990; 152:569–73. Shaikh N. Emergency management of fat embolism syndrome. J Emerg Trauma Shock 2009; 2:29–33.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
