This action might not be possible to undo. Are you sure you want to continue?
Luciano De Carlis,* Vincenzo Pirotta,* GianFranco Rondinara,* Cosimo V. Sansalone,* Giovanni Colella,* Giuseppe Maione,* Abdallah O. Slim,* Antonio Rampoldi,† Alberto Cazzulani,† Luca Belli,‡ and Domenico Forti*
Focal nodular hyperplasia (FNH) and adenoma are rare benign hepatic tumors, and the standards for diagnosis and treatment still remain controversial. Usually adenoma is an indication for resection, due to its tendency to bleed and to degenerate; FNH, on the contrary, may be treated conservatively. Preoperation differential diagnosis is, however, difficult, often impossible. Materials and methods. Thirty-eight patients with presumed hepatic adenoma and/or FNH were studied at our department from 1984 to 1996. Preoperative assessment included clinical evaluation and symptoms, laboratory tests, liver biopsy, ultrasound scan, computed tomography scan, magnetic resonance imaging, scintigraphy, and angiography. Thirteen patients had a presumed diagnosis of FNH, 16 of adenoma, and 9 of undetermined benign lesions; 27 had hepatic resections (3 with laparoscopic technique), and 11 were not operated on and are actually under a strict follow-up observation. Results. The ﬁnal diagnosis was 19 FNH and 19 adenomas (2 of which contained areas of hepatocarcinoma). Presumed diagnosis was conﬁrmed in 71% of cases. Use of oral contraceptives, abdominal symptoms, and pathologic liver test results were more frequent in patients with adenomas. There were no deaths after surgery. All resected patients were tumor free during the follow-up, and in 10 of the 11 nonoperated cases, the size of the nodules remained unchanged. We conclude that precise diagnosis of these benign liver tumors remains difficult and sometimes impossible, despite new imaging techniques. Hepatic resections can be performed under very safe conditions; laparoscopic surgery may play a role in selected cases. Adenomas and uncertain cases are clear indications for surgery. Only when a diagnosis of FNH can be ﬁrmly conﬁrmed in asymptomatic patients is strict observation without surgery recommended. Copyright r 1997 by the American Association for the Study of Liver Diseases
n contrast with hemangioma,1 focal nodular hyperplasia (FNH) and hepatic adenoma are very uncommon benign lesions affecting the liver, and their diagnosis and differentiation may be difficult. Moreover, their natural history is not well defined. Because of all these considerations, their surgical indication and treatment remain controversial. As a consequence of the widespread use of improved imaging modalities, these tumors are now recognized more frequently, and more information is available on their behavior.2 In particular, a
From the *Department of Surgery and Abdominal Transplantation, the †Department of Radiology, and the ‡Department of Hepatology, Niguarda Hospital, Milan, Italy. Address reprints request to Luciano De Carlis, MD, Divisione di Chirurgia Generale e dei Trapianti Addominali, ‘‘Pizzamiglio 27’’, Ospedale Niguarda, 20162 Milano, Italy. Copyright r 1997 by the American Association for the Study of Liver Diseases 1074-3022/97/0302-0009$3.00/0
strict correlation exists between these tumors and the use of oral contraceptives.3,4 Hepatic adenomas have the tendency to grow to conspicuous sizes, and spontaneous ruptures or bleeding are relatively frequent. Malignant degeneration has been reported in some cases, and resection is therefore advisable.5 On the contrary, FNH is often an incidental finding, and to date there is no convincing report showing that these tumors can bleed or degenerate. Because of this, resection may be avoided when the diagnostic assessment evidences FNH.2 In clinical practice, computed tomography (CT), magnetic resonance imaging (MRI), ultrasound (US), and angiography are used in an attempt to determine the nature of the solitary masses of the liver, but accurate distinction between adenoma and FNH before surgery is often difficult.6-10 Furthermore, percutaneous needle biopsy cannot differentiate these tumors with accuracy.11,12
Liver Transplantation and Surgery, Vol 3, No 2 (March), 1997: pp 160-165
46 months). whereas 3 patients had multiple tumors: Two had two FNH and 1 had three adenomas. The diagnosis was made by adopting predefined criteria. a differential diagnosis could not be obtained. gammaglutamyl transpeptidase. Nine patients were observed in the first 6 years. and segmentectomy or enucleation in 18. whereas the remaining 29 were referred to us between 1990 and 1996. Three superficial nodules. Ten (26. No chronic liver diseases nor abnormalities in serum alphafetoprotein levels were detected in any patients. when surgery was unadvisable. (Table 1). two located in the third and one in the sixth liver segment were excised by laparoscopic technique. and 9 with uncertain diagnosis underwent liver resection. including liver tests. (A) A hypodense area on the left lateral hepatic segments is present before contrast administration. 23 (60. 7 months to 12 years) before diagnosis. and in 7. FNH was preoperatively diagnosed in 13 cases and adenoma in 16.6). and red blood cell count showed some abnormalities in 14 cases (36. Nodules were solitary in 35 of 38 patients (92. in 8 of these cases no clinical symptoms were evident. and biochemical .6%). CT. All the lesions were submitted to extensive evaluation by a trained pathologist.5). including histology. available to us since 1990. Eleven patients were not operated on because preoperative study. and 15 (39.7 cm). slightly modified by the authors. All 11 of these lesions were the only lesion in each patient. Preoperative findings were matched with definite diagnoses and with the results of surgery.5 to 22 cm (mean. (B) A typical marked contrast enhancement is evident in the early arterial phase. showed the typical features of FNH. left lateral lobectomy in 4. 38 patients with either hepatic adenoma or FNH were observed in our surgical department. The aim of this study was both to define the diagnostic criteria and establish in which cases surgical treatment of these tumors is indicated.10. to determine the location of the tumor and its relationship with the vascular system. liver scintigraphy. MRI. Only alkaline phosphatase. Thirty women (78. 32. The patient population included 37 women and 1 man. In 9 patients. Follow-up was completed in 100% of cases and ranged from 2 months to 12 years (average.8 years (range. and selective hepatic angiography were performed in all cases. where fresh frozen section specimens were obtained during laparotomy. Patients underwent an annual check-up with clinical examination. 2). 16 with diagnosis of adenoma. which was acute in 9 (23.1%). ranging in age from 21 to 57 years (average age. Typical CT appearance of an adenoma of the left liver lobe. Right hepatectomy was performed in 3 cases.8%).3%) patients were completely asymptomatic.Hepatic Adenoma and FNH 161 The authors report herein their experience in the treatment of these benign lesions of the liver.9%) had a palpable mass. The size of the different nodules ranged from 2. An intraoperative US scan was used routinely Figure 1. 11 (28. left hepatectomy in 2. US scan.12 Percutaneous fine-needle liver biopsies were performed in all except 3 patients.4%) suffered from vague digestive troubles with fatigue and sense of heaviness in the right abdomen. All patients were evaluated with routine laboratory analyses. and the lesions were discovered during periodic routine examinations (8 cases) or laparotomies (2 cases) performed for different medical reasons. 1. with an average size at CT of 4. Two symptomatic patients with diagnosis of FNH.5 to 5.9%) had a history of oral contraceptive consumption for an average time of 5. the lesions were not easily resectable because of their central location in the liver parenchyma.5%) complained of abdominal pain.2 cm (range 2. was employed in the last 28 patients (Figs. US scan. the clinical courses of these patients were closely followed up through time. Intratumor hemorrhage was noted in five nodules. moreover.11. 8. Materials and Methods From January 1984 to May 1996.
the feeding artery is evident with rapid contrast ﬁlling from the center to the periphery of the node. and inﬂammatory cells Figure 2. an accurate pathological examination revealed areas of welldifferentiated hepatocarcinoma within the adenomatous nodules. vascular supply from the periphery to the center of the node At biopsy. calciﬁcations. Focal nodular hyperplasia of the left liver lobe. 23. ipodensity followed by a marked contrast enhancement. In one case the nodule size increased from 4. and/or histological features of FNH. Diagnostic Criteria for Adenoma and FNH Adenoma Enlarging nodules At CT. normal or increased uptake At biopsy. bile ducts. a central feeding artery with rapid visualization of the suprahepatic vein At MRI. sheets of normal hepatocytes without bile ducts and Kupffer cells FNH At CT.7 months. normal hepatocytes separated by ﬁbrous septa. Statistical analysis was based on the Student’s t-test. and fat inﬁltration At angiography. Preoperative biopsy and postop- data. The therapeutic protocol was approved by the ethical committee of the hospital. (B) At selective angiography a marked hypervascular lesion appears.5 to 5. Table 2. Two presumed adenomas were determined to be FNH. These *Including the 11 nonresected patients. range. In 2 patients.05. as mentioned earlier (average follow-up. The 11 patients with unresected FNH are regularly followed in our outpatient clinic. four were diagnosed as FNH and five as adenomas. capsule. †Including the 2 cases with areas of HCC. results appear in Table 2. Comparison Between Presumed and Deﬁnitive Diagnosis in the 38 Considered Patients FNH Presumed diagnosis Final diagnosis FNH Adenoma 13 13* 0 Adenoma 16 2 14† Uncertain 9 4 5 Results The final diagnosis in the resected cases was FNH in 8 patients and hepatic adenoma in 19. MRI. angiographic. CT and/or MRI were performed only when indicated. isointense lesion on T1–T2 with hyperintense central scar on T2 At scintigrams.5 cm and is now under strict evaluation. proliferating vessels. All 11 patients presented typical CT. assuming significance when P . . and the clinical courses were uneventful except for 1.162 De Carlis et al Table 1. . (A) MRI shows an isointense mass with a little hyperintense central scar on T2-weighted images. and an informed consent was obtained from each patient included in the study. All presumed FNH were confirmed both by pathological examinations and by follow-up data. 5 to 39). isodensity with an iperdense central scar (50% of cases) At angiography. whereas in the nine undetermined cases.
giving excellent diagnostic confirmation without any related complications. in some cases.3%) Vague 5 (26. when correctly diagnosed.9%) of FNH had a correct preoperative diagnosis with accurate imaging techniques.4%) Symptoms None 8 (42. Selective hepatic angiography was performed routinely in this series of patients. From our data. 6–25). All preoperative studies showed a diagnostic accuracy of 71% (27/38 patients). Oral contraceptive use was more frequent in the patients with adenoma (17/19 or 89.9 days (range. Furthermore. 3).05).1%) Abdominal pain 8 (42. the importance of angiography is incomparable for technical reasons when planning a liver resection.9 P Value ns ns ns ns .3%) Biochemical alterations 2 (10.2%). between benign and malignant tumors. In our experience.8 (21-43) 17 (89. and when used together. they may add 10% to 15% to specificity. P 5 ns). The 3 patients operated with the laparoscopic technique showed no postoperative problems and were discharged from hospital on the 4th postoperative day. in the typical cases. thus allowing a definite diagnosis in 19 of 35 cases (54. this observation is not reported in the literature and seems to be a pathognomonic picture of FNH. The preference for angiography results from our extensive experience in the treatment of portal hypertension. in which it proved to be extremely safe and exhaustive. No perioperative deaths occurred in the patients who underwent liver resection. Moreover.05 ns ns *Including two cases with areas of HCC..Hepatic Adenoma and FNH 163 erative surgical pathologic evaluation or follow-up data (in the nonoperated cases) were in agreement. Concern exists about its extensive utilization for benign hepatic lesions because it has the disadvantage of being an invasive procedure. only 15/19 (78. Other clinical features of our patient population are shown in Table 3. . CT permits diagnosis in typical cases when a central scar within the nodule or a feeding vessel can be observed. During the follow-up. Scintigraphy shows normal uptake in all cases of FNH due to the presence of Kupffer cells.3%) 6 (31. one patient died in a traffic accident 3 years after the resection. Three patients had subdiaphragmatic fluid collections: One was reoperated and a small biliary fistula was sealed. The average hospitalization time was 10. similar to the central scar described as typical for FNH.4 (23-57) Oral contraceptive use 13 (68. All the others are alive with no evidence of tumor recurrence. Minor complications occurred in 5 other patients including pleural effusion in 2.5%) Single lesion 19 (100%) Size (mean) 9. US scan is nonspecific in the differentiation of these lesions but has a great value as a noninvasive method in the follow-up of both resected and nonresected patients.4%) than in those with FNH (13/19 or 68. All patients had discontinued oral contraceptive use. may be managed conservatively and monitored with repeated US scans. no patient with adenoma or other hepatic masses evidenced such angiographic features. but these characteristic pictures are present only in 50% of patients.5%) 15 (78.2 Adenoma (n 5 19*) 31.14. but recent data Table 3.1%) Acute pain 0 (0%) Palpable mass 5 (26. To our knowledge. and in all. fibrolamellar carcinoma may present an important fibrotic component.8%) of our patients with FNH.5%) 10 (52. Discussion Our experience seems to confirm that FNH.6%) 12 (63. Acute pain (possibly related to intranodular bleeding) and pathologic liver test results were significantly more frequently associated with the presence of an adenoma or an hepatocarcinoma (P .4%) 2 (10.. Clinical Features of the Patients FNH (n 5 19) Age (mean 1 range) 33.13. pneumonia in 1. a suprahepatic vein selectively draining the mass was usually rapidly seen along with the feeding artery (Fig. All hemorrhagic nodules were adenomas. .9%) 9 (47.4%. MRI has an accuracy comparable to CT.05 ns ns . the other 2 maintained percutaneous drainages for a few days.15 Problems may exist in obtaining a certain differential diagnosis between FNH and adenoma and. this diagnosis was confirmed either by postoperative pathologic evaluation or by follow-up data.4%) 16 (84. and wound suppuration in 2: All were treated conservatively. The procedure has diagnostic value for FNH when a feeding artery to the mass is demonstrable: This was the case in 11 of 19 (57.2%) 7.
patients had symptoms present in a higher percentage of cases.7. Laboratory tests in most cases show alterations in stasis indexes. . but it undoubtedly seems that the incidence in patients using sex hormones who manifested FNH is higher than the percentage of Figure 3. there is a tendency to observe asymptomatic masses. 4).9. (A) Early arterial phase showing a hypervascular mass with a central feeding artery: The contrast dye rapidly ﬁlls the node from the center to the periphery. without any biochemical abnormalities. Typical angiographic images of focal nodular hyperplasia of the right liver lobe. Figure 4. contrastenhanced lesion is evident on the fourth liver segment. selectively draining the node. On scintigram a reduced uptake is usually evident. Percutaneous liver biopsy alone is reported to be of little value in the diagnosis of these benign tumors due to the frequent lack of specific features in a small specimen.11. is well evidenced. demonstrate normal uptake also in 25% of cases of adenoma.1%). CT and MRI frequently demonstrate the presence of necrosis or hemorrhage within the nodules (five cases in our series. especially when bleeding or sudden growth occurred13.14 (Fig. however. the material is often inadequate.8. and the distinction between adenoma and well-differentiated hepatocellular carcinoma remains difficult. Another nonhemorrhagic.6. In our series. Fig. and a suprahepatic vein. incidentally seen.164 De Carlis et al Adenoma was correctly diagnosed in 12 of 19 cases (63.10 Either an enlarging lesion or anemia on subsequent controls may indicate the presence of an adenoma.11 Our study confirms the strict correlation between adenoma and the use of oral contraceptives. however.12 Other problems are related to the fact that biopsy may be contraindicated in hemorrhagic lesions. 4). four of our cases confirmed this finding. these signs had statistical significance in the differentiation between adenoma and FNH. CT scan showed an enormous hemorrhagic and necrotic mass arising from the left liver lobe and occupying the whole left hypocondrium. moreover. such as large hepatomas. A 27-year-old female with double hepatic adenoma. but not always. (B) In late phases the node is completely opaciﬁed. these findings. In case of FNH. and typical signs were present in only 54. may be encountered also in malignant lesions.2% of our cases. less evident is the correlation in cases of FNH.13 Laboratory tests and symptoms are not diagnostic.
Liver cells adenomas: A 12 year surgical experience from a specialist hepatobiliary unit. The malignant transformation of liver cell adenomas. In: MacSween RNM.174:474-478. Belli L. Hepatic tumors: Possible relationship to use of oral contraceptives. McGill DB. 8. 5. Rondinara GF. In cases of undetermined diagnosis. Henderson B. Vasile N. and hepatic resections are now performed safely. Belghiti J. CT. Reddy KR. Vilgrain V. Brambilla G. Weiland LH. Hepatic adenoma and focal nodular hyperplasia: Clinical. Two patients in our series had adenomas containing degenerated areas of hepatocarcinoma. 12. Williams R. 19. Sheedy PF. Elective hepatic resection for benign and malignant liver disease: Early results. Laparoscopic excision of benign liver lesions. independent of their location and size.18 In conclusion. Panis Y. Charboneau JW. Liver-cell adenomas associated with use of oral contraceptives. in very selected cases (superficial ‘‘plongeant’’ lesions). Kerlin P. Sansalone CV. Skeel RT.Hepatic Adenoma and FNH 165 women in the general population in Italy using these drugs (68. 30%). Budin R.13:423-435. Davis GL. Hepatology 1990.184:699-703. 6. McGlynn F. Tumors and tumor like lesions of the liver and biliary tract. 17. Casarella WJ. Excisional therapy for benign hepatic lesions. Bennet WF. An important area on which to focus when studying these lesions is the risk of not identifying malignant tumors. Awad RW. Resection of presumed benign ´ ´ liver tumours. Vilgrain V. Mathieu D.80:1039-1041. Koh MK. be an operative option. Gastroenterology 1977. De Carlis L.131:393-402. in our opinion. Our philosophy. Review of hepatic imaging and a problem oriented approach to liver masses. Coombs RJ. Shiff ER. 14. Habib NA. Starzl TE. all the diagnostic preoperative studies in our series led to the right diagnosis in 71% of cases.9 days in our series). Iwatsuki S. Brown ML.14:44-47. Woldenberg LS. Malignant transformation of adenoma is a rare event. Semin Liver Dis 1993. 1987:574-645. FNH and liver cell adenoma: Radiologic and pathologic differentiation.78:956-958. Leese T. an interesting finding. In the last 10 years.294:470-472. Gastroenterology 1983. Caron-Pontreau C. 7. Arch Surg 1994. Meny Y. Bismuth H. Bottino G. Scheuer PJ (eds). Anthony PP. 4. Berman MM. Arrive L. Flejou JF. Approach to a liver mass. Focal nodular hyperplasia and hepatic adenoma: Comparison of angiography.73:386394. Beati C. therefore. Am J Roentgenol 1978. Klastin G. Bova JG.84:994-1002. Stephens DH. which specialized liver centers offer today. Pateron D. Johnson PM. Churchill Livingstone. Knowles DM. Br J Surg 1993. Pathology of the liver (ed 2). Focal nodular hyperplasia of the liver: MR imaging and pathologic correlation in 37 patients. Flejou JF. Asymptomatic patients who have a diagnosis of FNH based on the aforementioned typical signs are also under repeated clinical and ecographic controls. This fact may reflect a selection bias in the study but is. Surgical treatment of symptomatic 18. we usually resect all easily resectable lesions and keep under close observation those in which the risks of resection seem high. 2. Focal nodular hyperplasia of the liver: Results of treatment and options in management.32:524-527. Howard ER. giant hemangiomas of the liver. Obstet Gynecol 1991.19 The extensive use of hepatic resection in such cases can be justified by offering patients the guarantee of a higher recovery rate. Gimson AES.203:558-564.292: 1355-1357. but recent reports point out this possibility in an increasing percentage of cases. nevertheless the duration of the operation usually does not affect the recovery of patients in overall good condition. Foster JH. Anthony PP. Fekete F. Clin Imaging 1990. with low morbidity and very low mortality rates. Johnson TM. Farges O. Edinburgh.171: 240-246. References 1. 13. Gut 1991. In specialized surgical units these operations are done without any need of transfusions and with reduced hospitalization time. Najmark D. Adson MA. Drouillard J. Sheedy PF.80:380-383. 11. 16. . any increase in size or in imaging characteristics should be signal for excision. Ann Surg 1988.4% v. Hepatic adenomas and focal nodular hyperplasia: Dynamic CT study. Radiology 1992. N Engl J Med 1976.5. Surg Gynecol Obstet 1990. hepatic surgeons have largely improved their results. 10.12:761-775. 3. Radiology 1986. It is our opinion that this is the actual limit in the treatment of these tumors. Pain JA.156:593595. Wolf M. Zografos G. 15. De Caprio J. Belghiti J. with more than one quarter being misdiagnosed. et al. Bruneton JN. Welch TJ. Edmonson HA. Magnetic resonance imaging of hepatic adenoma.13-16 Laparoscopic surgery may. Merrick HW. Bishara HM. Br J Surg 1993. as well as fewer complications. is to resect all lesions preoperatively classified as adenoma. Benton B. Surg Gynecol Obstet 1992. Todo S. Benhamou JP.17 The prolonged duration of laparoscopic procedures compared with laparotomic techniques is still a major concern. 10.129:712-717. US and scintigraphy. Radiology 1985. Reich H. 9. pathologic and radiologic features. and hospitalization time is reduced (4 v. et al.