British Journal of Radiology (2004) 77, 911-916
© 2004 British Institute of Radiology
doi: 10.1259/bjr/56713551
Dysplastic nodules in liver cirrhosis: detection with triple phase helical dynamic CT
J H Lim, MD1,
M J Kim, MD1,
C K Park, MD2,
S S Kang, MD1,
W J Lee, MD1 and
H K Lim, MD1
Departments of 1 Radiology and Center for Imaging Science and 2 Diagnostic Pathology, Samsung Medical Center, Sungkyunkwan University School of Medicine, 50 Ilwon-dong, Kangnam-ku, Seoul, Korea 135-710
 |
Abstract
|
|---|
The purpose of this study was to determine the sensitivity of triple phase helical dynamic CT for detecting dysplastic nodules in patients with liver cirrhosis. 76 dysplastic nodules were confirmed by histopathological examination of the liver specimens after surgical resection in 21 patients or after explantation of the whole liver in 20 patients. Triple phase helical dynamic CT including arterial, portal venous and delayed phases was performed as a pre-operative evaluation for hepatocellular carcinoma. Two readers retrospectively evaluated the images. The presence of dysplastic nodules was determined by one-to-one correlation of the CT images and the pathological results in terms of the anatomical location and size of each nodule. Helical dynamic triple phase CT depicted eight of 76 dysplastic nodules (10%): five of 35 high grade dysplastic nodules (14%) and three of 41 low grade dysplastic nodules (7%). Triple phase helical dynamic CT is insensitive for detection of dysplastic nodules in cirrhotic livers.
|
Introduction
|
|---|
Dysplastic nodules, which are currently considered as pre-malignant lesions [1–4], are frequently found in cirrhotic livers, the incidence at autopsy or in explanted liver being 15–25% [5–7]. Patients with dysplastic nodules need close follow-up and may be considered for percutaneous ablation therapy. Detection of dysplastic nodules is known to be notoriously difficult using ultrasound, CT and MRI. The reported detectability is 0% by ultrasound [8], 38–39% on helical dynamic CT [7, 9] and 15% using MRI [10]. However, some of these reported accuracies were based on a small number of image-guided biopsy proven dysplastic nodules, which may not truly represent sensitivity or accuracy of detection.
Multiphasic dynamic helical CT has become one of the standard CT techniques for the detection and characterization of nodular lesions in cirrhotic liver parenchyma. However, there have been few studies evaluating the true detectability of dysplastic nodules in cirrhotic patients, based on resected or explanted liver specimens [7]. We undertook this study to assess the efficacy of triple phase helical dynamic CT for the detection of dysplastic nodules which were confirmed by histopathological examination following surgical resection or explantation of the liver.
|
Materials and methods
|
|---|
Patients
From February 1998 to February 2002, a total of 76 dysplastic nodules were histopathologically diagnosed in 41 patients who underwent hepatic resection surgery for hepatocellular carcinoma (n=21; 12 had hepatic lobectomy and 9 had segmentectomy) or liver transplantation for hepatocellular carcinoma or liver failure (n=20). 40 patients were proven to have liver cirrhosis associated with hepatitis B and one patient was proven to have alcoholic liver cirrhosis. Diagnosis of the dysplastic nodules was proved by histopathological examination of the resected liver specimen. All patients underwent triple phase helical dynamic CT scanning with a large dose bolus injection of contrast material prior to surgery. The time interval between CT examination and surgery was 0–72 days (mean 29.9 days). There were 37 men and four women, and their ages ranged from 36 years to 58 years (mean 52 years).
Triple phase helical dynamic CT
Triple phase helical scanning was performed on all of the 41 patients using HiSpeed Advantage helical scanners (General Electric Medical Systems, Milwaukee, WI). The scans were obtained through the liver in a craniocaudal direction with 7 mm collimation; with a 7 mm s–1 table speed (pitch 1.0) during a single breath-hold helical acquisition of 25–30 s, depending upon the size of the liver; and with a 7 mm reconstruction interval. For the triple phase CT scanning, the hepatic arterial phase, portal venous phase, and delayed phase were scanned 30 s, 60 s, and 180 s, respectively, after the start of the injection of 120 ml of non-ionic iodinated contrast materials, iopamidol (Iopamiro 300; Bracco, Milano, Italy) via the antecubital vein at a rate of 3 ml s–1 by power injector.
Image interpretation
Pre-operative triple phase CT scans were retrospectively reviewed on a 2000 x 2000 PACS (GE Medical Systems Intergrating Imaging Solutions, Mount Prospect, III) monitor by two radiologists (JHL, MJK) by consensus. Both wide and narrow window settings (window width 150 and window level 90) were used for the liver parenchyma and scrolling of stacked CT images was routinely used to pick up any nodular lesions. All nodular lesions, with either high or low attenuation not explainable by normal structures and different from the attenuation of the normal liver parenchyma, were interpreted as being potentially hepatocellular carcinoma or dysplastic nodules. Small wedge-shaped hyperenhancing lesions at the periphery of the liver on the arterial phase were regarded as pseudolesions and were thus excluded from the category of hepatic nodules. The numbers, sizes and locations of the nodular lesions were recorded. The criteria for a dysplastic nodule on helical CT were a nodule up to 2 cm in diameter [11] and showing distinct low attenuation during all three phases of helical CT or a nodule seen only on the delayed phase as low attenuation. The CT criteria for the diagnosis of hepatocellular carcinoma were a nodule showing hyperattenuation on the hepatic arterial phase with or without mosaic pattern, low attenuation or isoattenuating on the portal venous and delayed phases compared with the adjacent parenchyma. Hepatocellular carcinoma, hepatic cysts and haemangiomas were excluded by the coordinator (MJK) after initial interpretation for the study by comparing the CT images and other imaging in terms of locations and sizes. The diagnosis of cysts and haemangiomas was based on the typical CT findings in conjunction with at least one other imaging findings such as ultrasound or MRI, and pathological findings on the resected specimen or explanted liver.
Pathological examination
Resected liver specimens or explanted livers were serially sectioned in the transverse or coronal plane at 5–7 mm intervals (mean 6 mm) depending upon the anatomy of the resected liver specimen and the position of the hepatocellular carcinoma. Sectioning of the liver specimens was independently performed without information of CT findings. The liver specimens were examined in a fresh state immediately after surgery as well as in a fixed state in formalin by the same liver pathologist (CKP). All nodules suggestive of hepatocellular carcinoma or dysplastic nodule on the gross specimens were recorded and examined microscopically [12–14] by one liver pathologist (CKP).
Pathology–radiology correlation
All nodular lesions seen on triple phase CT were compared on a one-to-one basis by segment with corresponding hepatocellular carcinomas and dysplastic nodules in the resected or explanted liver specimens in terms of location, size and numbers. When an enhancing nodule which fell in the category of hepatocellular carcinoma matched a histopathologically confirmed hepatocellular carcinoma on a pathological specimen in terms of the size and anatomic location, the nodule was considered to be hepatocellular carcinoma (Figure 1
). When a non-enhancing nodule on CT images which fell in the category of dysplastic nodules correctly matched a histopathologically confirmed dysplastic nodules, the nodule was considered to be a dysplastic nodule (Figures 1 and 2). When findings on CT did not correlate with pathological examination, the pathological findings were considered the gold standard.
View larger version (96K):
[in this window]
[in a new window]
|
Figure 1. A 62-year-old woman with hepatitis B related cirrhosis who underwent right hepatic lobectomy for a hepatocellular carcinoma. Histopathological examination disclosed a 3.3 cm hepatocellular carcinoma (segment VIII) and eight low-grade dysplastic nodules measuring 5–11 mm. CT images on (a) arterial, (b) portal venous and (c) delayed phases shows an ill-defined, inhomogeneously enhancing hepatocellular carcinoma in the right hepatic lobe (segment VIII, large arrow) and a 5 mm low attenuating dysplastic nodule at the posterior part of the right lobe (segment VII, small arrow). The other seven dysplastic nodules were not depicted on CT. Note the heterogeneity of the hepatic parenchyma and the many small ill-defined nodules of subtle low attenuation representing cirrhotic regenerative nodules. Right and left hepatic bile ducts are dilated (arrowheads in b and c).
|
|
View larger version (81K):
[in this window]
[in a new window]
|
Figure 2. A 64-year-old man with hepatitis B related liver cirrhosis who underwent right hepatic lobectomy for a hepatocellular carcinoma in the segment VI. A high-grade dysplastic nodule was confirmed in the segment VII. CT images at (a) arterial, (b) portal and (c) delayed phase imaging show a discrete, low attenuating nodule representing a dysplastic nodule (thick arrow) in the resected specimen. Thin arrow points to a large regenerative nodule.
|
|
|
Results
|
|---|
A total of 76 dysplastic nodules were confirmed histopathologically in 41 patients. There were 35 high grade dysplastic nodules and 41 low grade dysplastic nodules, ranging in size from 5 mm to 20 mm. 22 patients each had 1 dysplastic nodule, 11 patients had 2 dysplastic nodules, 4 patients had 3 dysplastic nodules, 3 patients had 4 dysplastic nodules and 1 patient had 8 dysplastic nodules. There were 45 hepatocellular carcinomas in 41 patients.
On CT imaging, eight dysplastic nodules (10%, 8 of 76 dysplastic nodules) were depicted. Five of 35 high grade dysplastic nodules (14%) and three of 41 low grade dysplastic nodules (7%) were depicted and the difference is not statistically significant. The size of the dysplastic nodules depicted (average 1.2 cm, range 6–20 mm) was similar to that of the dysplastic nodules not visible on CT (average 1.0 cm, range 5–20 mm). The attenuation of the nodules was low-, low- and low- in arterial, portal venous and delayed phase, respectively, in three nodules (Figures 1 and 2
), iso- low-, and low-, in one nodule, iso-, iso-, and low- in three nodules, and high-, high-, and isoattenuating in one nodule (Figure 3). Seven dysplastic nodules were correctly categorized on CT (Figures 1 and 2) but one dysplastic nodule was categorized as hepatocellular carcinoma (Figure 3).
View larger version (104K):
[in this window]
[in a new window]
|
Figure 3. A 46-year-old man with hepatitis B related liver cirrhosis who underwent right hepatic lobectomy under the diagnosis of a suspected hepatocellular carcinoma in the segment VII. Histopathological examination disclosed a 14 mm low grade dysplastic nodule in the segment VII. CT images show an enhancing nodule (arrow) on (a) the hepatic arterial and (b) portal venous phases which is isoattenuating on (c) the delayed phase images; this represent the dysplastic nodule on histopathological examination.
|
|
Among the 45 pathologically proven hepatocellular carcinomas, 31 nodules were categorized as hepatocellular carcinomas and one hepatocellular carcinoma was categorized as dysplastic nodule on CT; 13 hepatocellular carcinomas were not depicted on CT. Thus, the sensitivity of CT for hepatocellular carcinomas was 71% (32 of 45 hepatocellular carcinomas). The detectabilities of dysplastic nodules and hepatocellular carcinomas were 16% (6/38) and 78% (21/27) in patients who underwent surgical resection, and 5% (2/38) and 56% (10/18) in patients who underwent liver transplantation, respectively.
There were multiple hepatic cysts ranging 1–2 cm in diameter in two patients, one cyst measuring 1.5 cm in one patient, one haemangioma measuring 2.5 cm in diameter; these benign lesions were confirmed at pathological examination.
|
Discussion
|
|---|
Few studies have described the detectability of dysplastic nodules on CT. Cho et al [9] reported a detection rate of 38% for dysplastic nodules on triple phase CT; the detection rates were 8% for small dysplastic nodules (<14 mm) and 76% in large dysplastic nodules (>15 mm), but the diagnosis of many dysplastic nodules in this study was based on the ultrasound guided biopsy. Hayashi et al [15] reported that 48% of dysplastic nodules were detected on CT during arterial portography and 47% were detected on CT during hepatic arteriography. This study described the detectability by CT during arterial portography and CT hepatic arteriography and half of the dysplastic nodules were diagnosed by ultrasound guided biopsy. It is accepted that the interpretation of biopsy specimens of dysplastic nodules is difficult on the basis of fragments of tissue from a biopsy needle. Therefore, the true detectability of dysplastic nodule in liver cirrhosis should preferably be based on the dysplastic nodules that are confirmed histopathologically by liver resection. Lim et al [7] reported that the detectability of dysplastic nodule was 39% with helical dynamic triphasic CT scanning in advanced cirrhotic patients who underwent liver transplantation. In their study, the detectability was based on 23 dysplastic nodules.
Our study showed that the detection of dysplastic nodules in cirrhotic liver on CT is very difficult, the sensitivity being only 10% (8 of 76 dysplastic nodules). There was no significant difference in the depiction rate depending on the size of the nodules or whether the nodules were high or low grade dysplastic nodules. This result implies that CT cannot be used to detect dysplastic nodules, although it is one of the standard methods for the diagnosis of hepatocellular carcinoma in patients with liver cirrhosis. The difference in detection sensitivity from our previous study [7] is probably due to the different histopathological criteria for dysplastic nodules. The criteria we used in the previous work was based upon large nodules greater than 0.5 cm in diameter that are distinct from the surrounding parenchyma in attenuation, texture or the degree to which they bulge from the cut surface of the liver. The criteria of dysplastic nodule in the current study was based on the criteria that was proposed by the International Working Party [12]. The definition of dysplastic nodules by the International Working Party is a nodular region of hepatocytes at least 1 mm in diameter with dysplasia but without definite histological criteria of malignancy. The criteria seem to be more strict: some nodules that were diagnosed as dysplastic nodules on the previous study [7] might have been large regenerative nodules by the criteria of the International Working Party. Therefore, some large regenerative nodules which were of high or low attenuation might have been detected on dynamic CT scans. Fat containing large regenerative nodules are typically of low attenuation [16].
It is not clear why only a small number of dysplastic nodules can be depicted on helical CT. The CT detectability of hepatic nodular lesions depends on the attenuation difference between the normal parenchyma and the nodular lesions. The vascular supply of any nodule on multiphasic dynamic helical CT depends on a complex mixture of hepatic arterial and portal blood supplies. Dysplastic nodules possess approximately the same amount of arterial supply as the surrounding hepatic parenchyma and a normal or slightly decreased portal venous supply [17, 18]. Therefore portal and delayed phase CT images may be better than arterial phase images in the detection of dysplastic nodules [19]. Capillarization and neoangiogenesis of dysplastic nodules is an ongoing process leading to different enhancement patterns [12, 20]. In about 10%, dysplastic nodules may possess a slightly increased abnormal arterial supply (through unpaired arteries) that compensates for a decreased normal arterial supply (through the portal tracts) to the nodules, but the portal venous flow is decreased [17–20]. Therefore, compared with surrounding liver most dysplastic nodules are isoattenuating or slightly low attenuating on arterial, portal and delayed phase images [19], and thus are difficult to depict on CT. In this regard, MRI may be superior to CT because of its ability to show internal characteristics using its excellent contrast resolution with various MR sequences and dynamic post-contrast imaging.
One of the factors which might affect the detectability of dysplastic nodules is the extent of cirrhosis [19]. In this study, detectability of dysplastic nodules in patients who underwent surgical resection was 16% whereas the detectability in patients who underwent explantation was 5%. The patients who had liver transplantation were very advanced cirrhotic patients (Child-Pugh class B or C), while the patients who had liver resection were patients with mild cirrhosis (Child-Pugh class A). The detectability of hepatocellular carcinomas appears to be dependant upon the degree of liver cirrhosis [7, 8, 10].
Our study has intrinsic weaknesses. First, accurate localization of a lesion in a Couinaud segment on CT, in the resected specimens or explanted liver, may be difficult in some cases because of the indistinct anatomical landmarks, difficult spatial orientation of the resected or explanted liver, and the shrunken and distorted anatomy of the cirrhotic liver. When there is more than one nodule on CT as well as in the resected liver specimen or explanted liver, it is difficult to correlate exactly which is which. Nonetheless, we attempted to be objective by determining the presence or absence of a nodule on CT without information about the pathology of the resected or explanted liver specimens and by correlating one-to-one for each nodule on the CT images and pathology.
Our study proved that dysplastic nodules are not readily detected on triple phase helical dynamic CT imaging, its sensitivity being 10%. Further study will be necessary for better detection using other contrast materials or other imaging modalities including MRI [19, 21]. Better detection sensitivity with MRI by using contrast-enhanced imaging has been reported [22].
|
Acknowledgments
|
|---|
We express our gratitude to Bonnie Hami of the University Hospitals of Cleveland, OH, for editorial assistance.
Received for publication January 27, 2004.
Revision received May 17, 2004.
Accepted for publication June 22, 2004.
|
References
|
|---|
- Nakanuma Y, Terada T, Ueda K, Terasaki S, Nonomura A, Matsui O. Adenomatous hyperplasia of the liver as a precancerous lesion. Liver 1993;13:1–9.[Medline]
- Arakawa M, Kage M, Sugihara S, Nakashima T, Suenaga M, Okuda K. Emergence of malignant lesions within an adenomatous hyperplastic nodule in a cirrhotic liver: observations in five cases. Gastroenterology 1986;91:198–208.[Medline]
- Sakamoto M, Hirohashi S, Shimosato Y. Early stages of multistep hepatocarcinogenesis: adenomatous hyperplasia and early hepatocellular carcinoma. Hum Pathol 1991;22:172–8.[CrossRef][Medline]
- Takayama T, Makuuchi M, Hirohashi S, et al. Malignant transformation of adenomatous hyperplasia to hepatocellular carcinoma. Lancet 1990;336:1150–3.[CrossRef][Medline]
- Terada T, Terasaki S, Nakanuma Y. A clinicopathologic study of adenomatous hyperplasia of the liver in 209 consecutive cirrhotic livers examined by autopsy. Cancer 1993;72:1551–6.[CrossRef][Medline]
- Theise ND, Schwartz M, Miller C, Thung SN. Macroregenerative nodules and hepatocellular carcinoma in forty-four sequential adult liver transplants with cirrhosis. Hepatology 1992;16:949–55.[Medline]
- Lim JH, Kim CK, Lee WJ, et al. Detection of hepatocellular carcinomas and dysplastic nodules in cirrhotic livers: accuracy of helical CT in transplant patients. AJR Am J Roentgenol 2000;175:693–8.[Abstract/Free Full Text]
- Kim CK, Lim JH, Lee WJ. Detection of hepatocellular carcinomas and dysplastic nodules in cirrhotic liver: accuracy of ultrasonography in transplant patients. J Ultrasound Med 2001;20:99–104.[Abstract]
- Cho JM, Kim SH, Lee WJ, et al. Adenomatous hyperplasia of the liver: three phase helical CT findings. J Korean Radiol Soc 1999;41:945–59.
- Krinsky GA, Lee VS, Theise ND, et al. Hepatocellular carcinoma and dysplastic nodules in patients with cirrhosis: prospective diagnosis with MR imaging and explantation correlation. Radiology 2001;219:445–54.[Abstract/Free Full Text]
- Efremidis SC, Hytiroglou P. The multistep process of hepatocarcinogenesis in cirrhosis with imaging correlation. Eur Radiol 2002;12:753–64.[CrossRef][Medline]
- International Working Party. Terminology of nodular hepatocellular lesions. Hepatology 1995;22:983–93.[CrossRef][Medline]
- Kojiro M. Hepatocellular nodular lesions. In: Pathology of early hepatocellular carcinoma and related lesions. Tokyo, Japan: Medical Publishers 1996:148–65.
- Theise ND. Macroregenerative (dysplastic) nodules and hepatocarcinogenesis: theoretical and clinical considerations. Semin Liver Disease 1996;15:360–71.
- Hayashi M, Matsui O, Ueda K, et al. Correlation between the blood supply and grade of malignancy of hepatocellular nodules associated with liver cirrhosis: evaluation by CT during intraarterial injection of contrast medium. AJR Am J Roentgenol 1999;172:969–76.[Abstract/Free Full Text]
- Choi D, Lim JH, Park CK, et al. Regenerative nodules in liver cirrhosis: sonographic appearance and pathologic correlation. J Korean Soc Med Ultrasound 1998;17:323–32.
- Ueda K, Terada T, Nakanuma Y, Matsui O. Vascular supply in adenomatous hyperplasia of the liver and hepatocellular carcinoma: a morphometric study. Hum Pathol 1992;23:619–26.[CrossRef][Medline]
- Matsui O, Kadoya M, Kameyama T, et al. Benign and malignant nodules in cirrhotic livers: distinction based on blood supply. Radiology 1991;178:493–7.[Abstract/Free Full Text]
- Lim JH, Choi BI. Dysplastic nodules in liver cirrhosis: imaging. Abdom Imaging 2002;27:117–28.[CrossRef][Medline]
- Park YN, Yang CP, Fernandez GJ, Cubukcu O, Thung SN, Theise ND. Neoangiogenesis and sinusoidal "capillarization" in dysplastic nodules of the liver. Am J Surg Path 1998;22:656–62.[CrossRef][Medline]
- Krinsky GA, Lee VS. MR imaging of cirrhotic nodules. Abdom Imaging 2000;25:471–82.[CrossRef][Medline]
- Ward J, Robinson PJ. How to detect hepatocellular carcinoma in cirrhosis. Eur Radiol 2002;12:2258–72.[Medline]
This article has been cited by other articles:
|
|
|
|
 
J.-J. Chung, J. S. Yu, J. H. Kim, M.-J. Kim, and K. W. Kim
Nonhypervascular Hypoattenuating Nodules Depicted on Either Portal or Equilibrium Phase Multiphasic CT Images in the Cirrhotic Liver
Am. J. Roentgenol.,
July 1, 2008;
191(1):
207 - 214.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
