The continuous development of highly sensitive clinical imaging increased the detection of focal lesions of the liver. These accidentally detected liver tumors without liver-specific symptoms such as cholestasis have been named “incidentalomas.” Diagnostic tools such as sonography, computed tomography, or magnetic resonance imaging are used increasingly in asymptomatic individuals without defined suspected diagnoses in the setting of general prevention or followup after a history of malignancy. But despite continuous improvement of diagnostics, some doubt regarding the benign or malign behavior of a tumor remains. In case an asymptomatic hemangioma or FNH can be preoperatively detected with certainty, the indication for surgery must be very strict. In case of symptomatic liver lesions surgical resection should only be indicated with tumor-specific symptoms. In the remaining cases of benign lesions of the liver, a “watch and wait” strategy is recommended. In case of uncertain diagnosis, especially in patients with positive history of a malignant tumor or the suspected diagnosis of hepatocellular adenoma, surgical resection is indicated. Due to the continuous improvement of surgical techniques, liver resection should be done in the laparoscopic technique. Laparoscopic surgery has lower morbidity and shorter hospitalization than open technique.
In recent years the rapid development of highly sensitive clinical imaging has led to the detection of focal lesions of the liver more frequently. In addition, diagnostic tools are used increasingly in asymptomatic individuals without defined suspected diagnoses in the setting of general prevention or followup after a history of malignancy [
Currently, there are no evidence-based guidelines regarding the appropriate approach to diagnosis, interpretation of imaging and laboratory findings, and the indication for surgical resection. Prospective, sufficiently powered, randomized controlled trials on the elective resection of benign liver lesions are lacking. Most recommendations are based on retrospective data with fewer than 60 patients or casuistic reports [
Generally a primarily conservative approach is considered to be the method of first choice in the treatment of proven benign liver tumors [
The cavernous and capillary hepatic hemangiomas, the focal nodular hyperplasia (FNH), and the hepatocellular adenoma are the most common benign lesions of the liver [
Two hemangiomas in gadolinium enhanced MRI (contrast medium: gadoxetic acid; disodium salt (Primovist, Eurokontrast GmbH, Heidelberg), scanner: GE Signa HDxt 1,5T (General Electric Company, USA)): peripheral nodular enhancement in T1 FS early arterial contrast phase (upper left): (T1 LAVA FS dynamic FA80, TR 185 TE 4,2), progressive centripetal enhancement in T1 FS late arterial (upper right): (T1 LAVA FS dynamic FA80, TR 185 TE 4,2) and portal-venous phase (lower left): (T1 LAVA FS dynamic FA80, TR 185 TE 4,2). Typical ill-shaped intermediate (less than in cysts) hyperintensity in T2 (lower right): (T2 FRFSE FS FA 90 TR 2500 TE 94,16). Lesion in left lobe is partially clotted with thrombosis and shows less enhancement.
These are usually diagnosed as asymptomatic incidental findings. In addition to nonspecific symptoms, hemangiomas also (rarely) rupture spontaneously or by trauma and then lead to acute hemorrhagic shock with upper abdominal pain [
Hemangiomas rarely occur in association with clinical syndromes. These include most of all the Kasabach-Merritt syndrome and the Blumgart-Bornman-Terblanche syndrome. The Kasabach-Merritt syndrome is characterized by a hemangioma bleeding, thrombocytopenia, and coagulopathy [
Hemangiomas generally have no growth tendency. In the literature, however, cases of hemangioma growth during pregnancy or after estrogen administration are described [
Treatment of choice is the parenchyma-saving enucleation or the sparing liver resection [
Focal nodular hyperplasia in gadolinium-enhanced MRI (contrast medium: gadoxetic acid; disodium salt (Primovist, Eurokontrast GmbH, Heidelberg), scanner: GE Signa HDxt 1,5T (General Electric Company, USA)): inhomogeneous hyperintensity on T1 FS in portal-venous contrast phase (T1 FSPGR FS FA12, TR 4,24 TE 2,04, TI 7).
10–20% of FNHs occur multifocally [
In case of a FNH, there is primarily no indication for surgical intervention [
Segment 1 adenoma in gadolinium-enhanced MRI (contrast medium: gadoxetic acid; disodium salt (Primovist, Eurokontrast GmbH, Heidelberg), scanner: GE Signa HDxt 1,5T (General Electric Company, USA)): hyperintens. in T1 FS arterial contrast phase (upper left): (T1 LAVA FS dynamic FA80, TR 185 TE 4,2), partial equilibration to liver isointensity in T1 FS late portal-venous phase (upper right): (T1 FSPGR FS FA12, TR 4,24 TE 2,04, TI 7), slight hyperintensity on T2 FS (lower left): (T2 FRFSE FS FA 90 TR 2500 TE 94,16), and isointensity in unenhanced T1 fat sat. (lower right): (T1 LAVA FS dynamic FA80, TR 185 TE 4,2).
This lesion of the liver is often diagnosed as an incidental finding in asymptomatic patients. In association with an adenoma often right-sided upper abdominal pain (80%) with normal liver function values occurs [
Beside the risk of acute bleeding complications, hepatocellular adenomas have a malignant degeneration risk from 4.2 to 10%, especially in inflammatory adenomas on MRI and/or the possibility of beta-catenin expression [
The adenomatosis, with more than 10 adenomas in an otherwise normal liver, is a special form. This is gender unspecific and has no association with the use of hormones [
The diagnostic differentiation from hemangioma is straightforward [
Atypical tumors of the liver, such as the angiolipoma or cystadenoma, often have an inhomogeneous structure that usually a precise preoperative classification of a benign or a malignant tumor is impossible. This is the reason why in case of these tumors oncologic resections with appropriate security clearance are recommended [
The conservative and surgical treatment of benign lesions of the liver includes several regenerative or real-neoplastic tumor entities. Depending on their origin, these tumors are divided in hepatocellular, endothelial, biliary, mesenchymal, and connective tissue tumors [
Surgically relevant tumor entities [
Pseudotumors* | Benign neoplasia | |
---|---|---|
Hepatocellular tumors | Focal nodular hyperplasia (FNH) | Hepatocellular adenoma |
Endothelial tumors | Hemangioma | |
Biliary tumors | Von Meyenburg complex | Biliary cystadenoma |
Biliary duct adenoma | ||
Mesenchymal tumors | Hamartoma | |
Connective tissue tumors | Lipoma, angiolipoma, fibroma, leiomyoma | |
Mixed-cellular tumors | Teratoma |
The benign lesions of the liver can be divided in solid tumors, in tumors with solid areas, or cystic tumors. They are uni- or multilocular [
As illustrated in Table
Tumour-associated demographics [
Prevalence | Age | F: M | Location | Size | Specialties | |
---|---|---|---|---|---|---|
Hemangioma | 5–20% | 35–65 | 2–6 : 1 | Subcapsular 90% |
<5–30 cm | Synchronic hemangioma in skin, lung, or brain (10–15%); |
| ||||||
FNH | 2-3% | 30–50 | 8 : 1 | Subcapsular 80% |
<5–15 cm | Growing: association with OC; rarely clin. symptoms |
| ||||||
Adenoma | Rare |
25–45 | 10 : 1 | Subcapsular |
5–15 cm |
Arise and growth: association with OC (>5 years), |
*a: year, OC: oral contraceptive, DIC: disseminated intravascular coagulopathy.
Clinical symptoms of the patient are crucial for the extent and type of diagnostic measures to be executed. If the patient is burdened by severe symptoms of the tumor, resection is indicated and further diagnostic workup is dispensable. On the other hand, with an asymptomatic tumor and nonspecific findings, every attempt has to be made to ensure the diagnosis. The contrast medium- (CM-) based computed tomography (CT) (multiphase spiral CT) and magnetic resonance imaging (MRI), especially when using liver-specific CM (e.g., gadoxetic acid (Primovist)), are the methods of choice in the diagnosis of benign liver tumors [
In clinical routine, sonography is primarily used as screening method. Through the use of CM and technical enhancements, such as tissue harmonic imaging, the importance of sonography in the diagnosis of benign liver lesions has greatly increased. The disadvantage of this method of investigation is the investigator dependency [
Even nuclear medical procedures, such as the erythrocyte pool scintigraphy with 99Tc or hepatobiliary scintigraphy under the utilization of tumor-specific characteristics without precise morphological mapping of the lesion, can achieve a high sensitivity in diagnostics. Relatively high costs and limited availability preclude nuclear medicine procedures from more extensive use despite their proven diagnostic value [
Morphology of the most common benign lesions in imaging techniques [
Ultrasonography | Triphasic CT | MRI | 18F-FDG PET scan | CT angiography | |
---|---|---|---|---|---|
Hemangioma (Figure |
More often: cavernous (high flow): heterogeneous, hypoechoic, sometimes calcifying |
Early phase: iridic diaphragm phenomenon with peripheral nodular enhancement |
Peripheral enhancement, centripetal progression, |
No uptake or photopenic defect compared to liver baseline | Cotton wool pooling of contrast, |
| |||||
FNH |
Homogeneous, iso-, hypo- or hyperechoic, |
Isodense with liver, |
Native: isodense |
No uptake | Hypervascular 70%; |
| |||||
Adenoma (Figure |
Unspecific, |
Homogenous > heterogeneous |
T1 Gd: hyperintense T2: hyperintense |
No uptake |
Hypervascular; |
CT: computed tomography, MRI: magnetic resonance imaging, FNH: focal-nodular hyperplasia CM: contrast medium, HCC: hepatocellular carcinoma.
In medical literature there are only few publications with larger numbers of cases that match radiological diagnosis with the corresponding histopathologic findings after surgical resection. Grimm et al. showed, in 26 cases of histological confirmed benign liver tumors, that the multiphase CT or MRI examination only in 54% of the cases produced the correct preoperative diagnosis [
For the highest diagnostic “security” preoperatively, at least one imaging procedure should be performed using a suitable, liver-specific CM in combination with a nuclear medicine examination [
In addition to liver enzymes, bilirubin, and cholestasis, the laboratory testing should include the determination of the tumor markers AFP, CA 19-9, and CEA to distinguish it from malignant tumors of the liver [
In case of a symptomatic tumor, determinants of resectability such as size, location, relationship to the hilum, and the blood vessels are in the focus. Thereby a minimum of diagnostics, such as a sectional imaging, is sufficient. If in case of an asymptomatic neoplasia surgery is necessary, a graduated diagnostic procedure should be performed to determine the exact type of tumor and the subsequent appropriate therapy [
In the literature, the performance of a percutaneous fine-needle biopsy (FNB) is controversially discussed. Because of the insufficient validity and often missing therapeutic consequences, the FNB should not be performed [
In clinical practice preoperatively, the exact morphology of a tumor in up to 35–45% of cases is not clearly determined by radiological investigations. With the suspected diagnosis of a possible malignancy in these cases, often and completely understandable surgical resection of the lesion is indicated [
In case of a suspected adenoma due to the possible risk of malignant transformation or occult malignancy, surgical resection should be done initially or rarely after transarterial embolization (TAE). In acute bleeding with hemorrhagic shock, the extent of the diagnosis depends on the one hand on the urgency of the operation and on the other if the entity of tumor is previously known. In case of a known hemorrhagic hemangioma instead of a risky emergency surgery, a therapeutic TAE could be tried [
Because of preventive checkups, followup, and screening examinations, increasingly more asymptomatic liver lesions, especially in patients with a positive history of malignant tumor, are newly diagnosed. This has led to a shift in the spectrum of indications for surgery from symptomatic to asymptomatic patients. Depending on the tumor localization, specific symptoms such as a laboratory chemical cholestasis or portal hypertension are possible [
The indication for surgery should be primary found in relation to the symptoms and the suspected diagnosis. In descending order in cases of an acute symptomatology such as bleeding, the suspicion of an adenoma or a lesion of unknown dignity, in particular with positive tumor history, a tumor-induced symptomatology, the progression of size, some nonspecific symptoms, or eventually also in case of a cancerophobia of the patient liver resection should be indicated [
In case of a prior history of hemangioma, if a spontaneous or traumatic rupture happens, the indications are entirely provided for the conservative approach, eventually after successful TAE [
The recommendations for surgical management of benign liver lesions are based on the results of retrospective analysis or case reports with fewer than 60 patients (medical evidence 3-4) [
If the indication for surgery was found, parenchyma sparing techniques for operation should be preferred. If the diagnoses of hemangioma or FNH are proven, enucleation is the method of choice. With the enucleation the loss of functional liver parenchyma and, at the same time, blood loss and the risk of bile leaks can be reduced [
Waterjet dissector (Helix Hydro-Jet; Erbe Elektromedizin GmbH, Tübingen, Germany).
This study only mentioned patients with one of the three most popular benign liver lesions, the hemangioma, the hepatocellular adenoma, and the FNH. Not for all patients of our cohort, every diagnostic possibility, named in Table
At the time of the liver resection, the age of the subjects in our series ranged from 23 to 72 years with a mean of 52.8 years. The ratio of men to females was 1.7 : 1. Tables
Mean age in years of the subjects of our own collective and the literature.
Dachau | Zülke et al. | Charny et al. | Weimann et al. | |
---|---|---|---|---|
[ |
[ |
[ | ||
Hemangioma | 54.9 | 49.5 | 52 | 47.6 |
FNH | 54.3 | 36 | 38 | 35.3 |
Adenoma | 52.3 | 40 | 34 | 34 |
Gender distribution (females: men) of the subjects of our own collective and the literature.
Dachau | Zülke et al. [ |
Charny et al. [ |
Weimann et al. [ |
Skalický et al. [ | |
---|---|---|---|---|---|
Hemangioma | 1.6 : 1 | 1.2 : 1 | 3 : 1 | 2.9 : 1 | 2.3 : 1 |
FNH | 2 : 1 | 2.5 : 1 | 9.5 : 1 | 11.5 : 1 | 2.1 : 1 |
Adenoma | 1.3 : 1 | 2.4 : 1 | 5 : 1 | 3.9 : 1 | 1 : 2 |
In patients with asymptomatic liver lesions, surgical intervention was mainly indicated since a primary or secondary malignancy could not be excluded with certainty. In case of symptomatic liver lesions, the indication for surgery was conducted on the one hand because of clinical symptoms, but on the other because a primary or secondary malignancy could not be preoperatively excluded with certainty by means of morphological imaging. Most often patients of our series complained about right upper abdominal pain. Our patients also reported about indigestion, loss of appetite, and icterus. Table
Indications for surgery of the three most common benign liver tumors (multiple answers in our own collective and Charny et al. [
Number of patients ( |
Health complaints (%) | Moot malignancy (%) | |||||||
---|---|---|---|---|---|---|---|---|---|
Dachau | Zülke et al. | Charny et al. | Dachau | Zülke et al. | Charny et al. | Dachau | Zülke et al. | Charny et al. | |
[ |
[ |
[ |
[ |
[ |
[ | ||||
Hemangioma | 21 | 12 | 39 | 29 | 58 | 59 | 71 | 25 | 33 |
FNH | 12 | 21 | 18 | 17 | 42 | 44 | 92 | 33 | 61 |
Adenoma | 7 | 15 | 8 | 88 | 33 | 37 | 57 | 67 | 75 |
Beside the clinical and radiological diagnosis, the tumor location and size determine the extent and type of surgery. Table
Distribution of surgical procedures.
Dachau | Zülke et al. [ |
Charny et al. [ |
Weimann et al. [ | |
---|---|---|---|---|
|
|
|
| |
Atypical resection | 20 (46) | 15 (28) | 33 (48) | 84 (50) |
Segmental resection, resection of more than one segment | 19 (44) | 32 (60) | 12 (18) | 38 (22) |
Hemihepatectomy | 4 (9) | 6 (11) | 23 (34) | 47 (28) |
| ||||
Total | 43 | 53 | 68 | 169 |
The distribution of the respective applied surgical procedure shows in our collective of subjects as well as the collective of the reference centers a selected group. Therefore, these results do not reflect the expected approach in the surgical treatment of benign liver lesions. Normally in the vast majority of cases, atypical resections or enucleations would be expected and less segmental resections, resections of more than one segment, or hemihepatectomies.
Table
Distribution of tumor-specific surgical procedure in our collective of subjects.
Atypical resection | Segmental resection | Resection of more than one segment | Hemihepatectomy | Total | |
---|---|---|---|---|---|
Hemangioma | 11 | 6 | 2 | 2 | 21 |
FNH | 6 | 3 | 1 | 2 | 12 |
Adenoma | 2 | 4 | 1 | 0 | 7 |
Others | 1 | 1 | 1 | 0 | 3 |
The low mortality rates result from the generally good preoperative liver morphology and function in patients with benign liver lesions. The rate of postoperative complications (minor) in our collective amounts in total for more than 15%. These results emphasize the need of a strict indication for liver resection, especially in primary asymptomatic patients (Table
Mortality and morbidity of our subjects after resection of benign liver tumors in comparison to the literature.
Mortality | Morbidity* | Revision | Hospitalization** | |
---|---|---|---|---|
Weimann et al. [ |
0.6% (1/173) | 24.9% (43/173) | — | — |
Charny et al. [ |
0% (0/68) | 20.6% (14/68) | — | 8.5 d |
Petri et al. [ |
0.9% (1/113) | 27.4% (31/113) | — | — |
Zülke et al. [ |
0% (0/55) | 18.5% (10/55) | <5% | 10 d |
Dachau | 0% (0/43) | 16.3% (7/43) | 0% | 11 d |
The mortality and morbidity of laparoscopic liver resections in our own collective are presented in Table
Mortality, morbidity, and hospitalization after resection of benign liver tumors during laparoscopic liver resections in our collective and share of laparoscopic procedures in relation to the total number of liver resections.
Mortality | Morbidity* | Hospitalization** | Distribution*** | |
---|---|---|---|---|
Hemangioma | 0% (0/7) | 14.3% (1/7) | 7 d | 33.3% |
FNH | 0% (0/8) | 0% (0/8) | 11 d | 66.7% |
Adenoma | 0% (0/4) | 0% (0/4) | 9 d | 57.1% |
| ||||
Total | 0% (0/19) | 5% (1/19) | 9 d | 47.5% |
Table
Number of patients with a history of malignant tumor disease where hepatic resection was performed.
Positive history for carcinoma | Radiological, no safe exclusion of a metastasis | |
---|---|---|
Hemangioma | 13 | 12 |
FNH | 4 | 3 |
Adenoma | 2 | 1 |
Table
Consistency of the preoperative diagnosis with the final postoperative histology.
Dachau | Charny et al. [ |
Zülke et al. [ | |
---|---|---|---|
Hemangioma | 7/16 (44) | 27/39 (69) | 9/12 (75) |
FNH | 4/12 (33) | 7/18 (39) | 10/21 (48) |
Adenoma | 3/7 (43) | 6/8 (75) | 5/16 (31) |
Preoperatively unknown | 5 | — | — |
In 12% of cases a lesion of the liver was not known preoperatively and noticed during a laparoscopic cholecystectomy. In order to get histology, this incidental findings were completely laparoscopically, atypically resected. The subsequent histological examination showed uniformly the diagnosis of a hemangioma. In patients with a positive history of malignant disease, every intrahepatic mass of uncertain dignity was resected. Our high rate of 56% preoperative misdiagnosis for the hemangioma is therefore the result of including patients without a preoperative diagnosis and patients with limited preoperative imaging. If patients with a history of carcinoma or incidental perioperative findings are not considered in the analysis, there is an acceptable agreement of 76% of the results. With regard to FHN and adenoma, preoperative diagnosis was primarily uncertain or unreliable, this being particularly relevant forFNH with 58% and for adenoma with 72% when taking into account the patients with a history of cancer.
The continuous development of imaging techniques enables to diagnose asymptomatic liver lesions with an increasingly high sensitivity and specificity [
In patients without history of malignant tumor, the FNH and the hemangioma are relatively straightforward to diagnose [
Algorithm for management of solid liver lesions (mod. from Terkivatan et al. [
Initially often close monitoring of liver lesions occurs without any major risk to the patient [
The recommendations for surgical management of benign liver lesions are based on the results of retrospective analysis or case reports with fewer than 60 patients (medical evidence 3-4) [
At diagnosis of hepatocellular adenoma, surgical resection is indicated, because of the potential risk of malignant transformation and a possible life-threatening bleeding complication, even in case of a definite tumor regression and discontinuation of oral contraceptives [
Necrosis, hemorrhage, and thrombosis of benign lesions of the liver may often complicate the diagnosis [
The implementation of a percutaneous fine-needle biopsy (FNB) is discussed controversially in the literature. Due to the low expressiveness and lack of therapeutic consequences in most cases, FNB should not be performed [
Asymptomatic hemangiomas do not require therapy [
At the diagnosis of FNH, primarily no surgical intervention is indicated [
Requirements for a surgical intervention of benign lesions [ low surgical mortality (<1%), low morbidity, avoidance of blood transfusion, good long-term results.
Our study and the current literature revealed that these conditions are fulfilled (see Section
The postoperative morbidity in our collective in total was 16.3%, whereas no major complications occurred and no revision surgery had to be performed. This result of the morbidity for liver resections at benign liver lesions is generally equivalent to medical literature (7–28%) [
In case of symptomatic benign tumors of the liver, surgical intervention in 80% of cases leads to a decrease of complaints [
Medical literature offers a low postoperative complication rate, in laparoscopic and open liver resections [
Despite continuous improvement of diagnostic possibilities, the dignity for up to 40% of incidentally detected liver lesions cannot be determined reliably till the final postoperative histology. In case of uncertain diagnosis, especially in patients with positive history of a malignant tumor or the suspected diagnosis of hepatocellular adenoma, surgical resection is indicated. In case an asymptomatic hemangioma or FNH can be preoperatively detected with certainty the indication for surgery must be very reluctant. In case of symptomatic liver lesions, surgical resection should only be indicated with tumor-specific symptoms. In the remaining cases of benign lesions of the liver, a “watch and wait” strategy is recommended. Due to the continuous improvement of surgical techniques, liver resection should also be done in the laparoscopic technique in case of more than one liver-segment resection or hemihepatectomy. Laparoscopic surgery has lower morbidity and shorter hospitalization than open technique.