SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Disclosure
  8. References
  9. Supporting Information

Background:

The incidence of metabolic syndrome-associated hepatocellular carcinoma (MS-HCC) is increasing. However, the results following liver resection in this context have not been described in detail.

Methods:

Data for all patients with metabolic syndrome as a unique risk factor for HCC who underwent liver resection between 2000 and 2011 were retrieved retrospectively from an institutional database. Pathological analysis of the underlying parenchyma included fibrosis and non-alcoholic fatty liver disease activity score. Patients were classified as having normal or abnormal underlying parenchyma. Their characteristics and outcomes were compared.

Results:

A total of 560 resections for HCC were performed in the study interval. Sixty-two patients with metabolic syndrome, of median age 70 (range 50–84) years, underwent curative hepatectomy for HCC, including 32 major resections (52 per cent). Normal underlying parenchyma was present in 24 patients (39 per cent). The proportion of resected HCCs labelled as MS-HCC accounted for more than 15 per cent of the entire HCC population in more recent years. Mortality and major morbidity rates were 11 and 58 per cent respectively. Compared with patients with normal underlying liver, patients with abnormal liver had increased rates of mortality (0 versus 18 per cent; P = 0·026) and major complications (13 versus 42 per cent; P = 0·010). In multivariable analysis, a non-severely fibrotic yet abnormal underlying parenchyma was a risk factor for major complications (hazard ratio 5·66, 95 per cent confidence interval 1·21 to 26·52; P = 0·028). The 3-year overall and disease-free survival rates were 75 and 70 per cent respectively, and were not influenced by the underlying parenchyma.

Conclusion:

HCC in patients with metabolic syndrome is becoming more common. Liver resection is appropriate but carries a high risk, even in the absence of severe fibrosis. Favourable long-term outcomes justify refinements in the perioperative management of these patients. Copyright © 2012 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Disclosure
  8. References
  9. Supporting Information

Non-alcoholic fatty liver disease (NAFLD) has emerged as one of the most frequent forms of chronic liver disease in developed countries1. It is closely related to insulin resistance, the clinical and biological manifestations of which are included in the metabolic syndrome. NAFLD encompasses a wide spectrum of diseases ranging from simple steatosis to non-alcoholic steatohepatitis (NASH), and can progress to cirrhosis. Any of these pathological conditions may predispose to the development of hepatocellular carcinoma (HCC)2–6. In particular, recent evidence suggests that 30–60 per cent of patients with metabolic syndrome develop HCC without severe underlying fibrosis2–6. In the absence of symptomatic liver disease it appears that many of these patients present at a more advanced age with large tumours at the time of diagnosis than patients with chronic liver disease of other aetiology4. As a consequence, few therapeutic options other than liver resection are left in this context.

Operative risk according to the severity of chronic liver disease and prognostic factors for longer survival, such as tumour size, differentiation and vascular invasion, have been studied extensively in the context of liver resection for HCC7, 8. These issues are still largely unclear in the setting of patients with HCC developed in a background of metabolic syndrome. Both impaired performance status of these patients as a consequence of advanced age, central obesity, diabetes and cardiovascular co-morbidities, as well as the impact of the underlying liver status, could adversely affect the postoperative course. The frequent absence of severe underlying fibrosis may result in these patients being considered as having a ‘normal’ liver. This, in turn, may result in inadequate preparation for liver resection9–11.

Only one small series has focused on the outcomes of surgical treatment of HCC in patients with NAFLD6, but that study did not specifically address the presence of an associated metabolic syndrome. Thus, the aim of the present study was to characterize the outcomes of liver resection in patients who developed HCC in a background of metabolic syndrome (MS-HCC) in order to determine the influence of the underlying liver on both the postoperative course and long-term outcome.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Disclosure
  8. References
  9. Supporting Information

Study design and study population

From a prospectively collected database of all patients who underwent partial liver resection for HCC between 2000 and 2011 at Beaujon Hospital, Clichy, France, data for all patients with metabolic syndrome as a unique risk factor for HCC were retrieved retrospectively. The diagnosis of MS-HCC was considered only in patients with the metabolic syndrome and after exclusion of other usual aetiologies of HCC, such as chronic hepatitis B or C infection, excessive alcohol consumption, haemochromatosis and autoimmune liver disease. Patients with an incomplete diagnostic workup and those with metabolic syndrome associated with another predisposing factor for HCC were not included. The diagnosis of metabolic syndrome included insulin resistance as the essential component and was considered when three or more of the following criteria were present12: central obesity; dyslipidaemia (triglycerides 1·7 mmol/l or above, or high-density lipoprotein cholesterol less than 1·03 mmol/l in men or less than 1·29 mmol/l in women); type II diabetes or glucose intolerance with fasting glucose 5·6 mmol/l or above; and hypertension (blood pressure above 135/85 mmHg). Because of the retrospective nature of the study, it was assumed that central obesity was reached when the patient's body mass index was greater than 28 kg/m2, as described previously1. Similarly, it was assumed that patients receiving statin or fenofibrate medication had dyslipidaemia.

Preoperative evaluation and postoperative outcomes

Preoperative investigations included complete blood and liver function tests as well as routine cardiorespiratory evaluation. Computed tomography and magnetic resonance imaging during the later years of the study were performed to assess both underlying liver and tumour characteristics. Initially, preoperative percutaneous biopsy of the non-tumorous parenchyma was performed on a case-by-case basis, but was carried out routinely during the final 5 years of the study. When severe fibrosis or cirrhosis was suspected, patients considered for major resection underwent preoperative transarterial chemoembolization (TACE) of the tumour, followed 4–6 weeks later by portal vein embolization (PVE)13.

All resections were performed with curative intent. Major liver resection was defined as resection of three or more Couinaud segments. All but three procedures were performed via a laparotomy. For all procedures, liver transection was performed with the crush-clamp technique or ultrasonic dissection under low central venous pressure (less than 5 mmHg), as described previously10. Haemostasis and biliostasis were achieved using either bipolar cautery coagulation or ligation for small pedicles. Intermittent pedicle clamping was performed in case of bleeding, or routinely in some patients in order to obtain a bloodless operative field. Methylene blue injection through the cystic drain was performed routinely to rule out biliary leakage. All intraoperative parameters, including type and duration of vascular clamping, blood loss with subsequent blood transfusion and duration of surgery, were recorded.

Postoperative complications were stratified according to the Clavien–Dindo classification14, which defines major complications by a score of 3 or more. Specific liver complications, encountered more often after major liver procedures, were detailed as follows: liver failure was defined according to the ‘50–50 criteria’ on postoperative day 515; ascites was defined as an abdominal drainage output of more than 10 ml per kg per day after the third postoperative day16; and biliary leakage was defined by a bilirubin concentration in the drainage fluid more than threefold higher than that in serum17. Both complications and operative mortality were considered as those occurring within 90 days of surgery, or at any time during the postoperative hospital stay.

Histological analysis

In every patient, preoperative biopsies and resected liver specimens were reviewed by a single pathologist specialized in liver disease and liver cancer. The pathological findings for some of the patients have been detailed previously4. Tumour characteristics, including size, number of nodules, presence of a capsule, satellite nodules, vascular invasion, and grade of differentiation, were assessed. Liver parenchyma was assessed according to the presence of: fibrosis, staged from 0 to 4, and considered severe in patients with stages F3 and F4; and NAFLD activity score (NAS), graded from 0 to 7 including steatosis (on a scale of 0–3), lobular inflammation and hepatocellular ballooning (each on a scale of 0–2)18. The diagnosis of NASH was made in patients with a NAS of 4 or more, or whenever mentioned by the pathologist, as suggested recently19. Liver parenchyma was considered normal in the absence of severe fibrosis and with a NAS below 2 (corresponding to maximal steatosis of less than 33 per cent without inflammation). Liver parenchyma was considered abnormal either in patients with severe underlying fibrosis (grade F3 or F4) and in those with without severe underlying fibrosis but with a NAS of 2 or more.

Statistical analysis

Patient baseline characteristics are expressed as median (range) for continuous data, and as numbers with percentages for categorical data. Preoperative, operative and postoperative characteristics, as well as long-term survival of patients, were compared according to the presence of parenchymal changes in the underlying liver (normal versus abnormal). Fisher's exact test was used to compare differences in categorical variables, and the Wilcoxon rank sum test for continuous variables. Variables achieving statistical significance at the 0·1 level in univariable analysis were considered for multivariable analysis. A backward variable procedure was used to identify independent predictive factors. Given the small number of events, multivariable analysis of risk factors for mortality was not conducted. Overall survival was defined as time from surgery to death (all causes). Disease-free survival was defined as time from surgery to first recorded evidence of recurrence on imaging. Both overall and disease-free survival were considered on an intention-to-treat basis, and therefore included in-hospital deaths. Cumulative overall survival and disease-free survival rates were determined using the Kaplan–Meier method and compared with the log rank test. All tests were two-sided. P < 0·050 was considered statistically significant. Statistical analyses were performed using SPSS® version 18·0 for Windows® software (IBM, Armonk, New York, USA).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Disclosure
  8. References
  9. Supporting Information

A total of 560 patients underwent liver resection for HCC during the study period. Of these, 62 patients (11·1 per cent) had the metabolic syndrome as a unique risk factor for HCC, including 24 (39 per cent) with a normal and 38 (61 per cent) with an abnormal underlying liver. During the study interval, the proportion of resected HCCs labelled as MS-HCC among patients undergoing liver resection for HCC increased continuously, from 2·5 per cent in the early part of the study to more than 15 per cent of all resected HCCs in more recent years (Fig. 1).

thumbnail image

Figure 1. Evolution of the proportion of resected hepatocellular carcinomas (HCCs) associated with the metabolic syndrome in the past decade

Download figure to PowerPoint

Patient characteristics

The study population comprised a high proportion of men (58 male patients, 94 per cent) and elderly patients, including 27 patients (44 per cent) aged 70 years or more. The median number of metabolic syndrome factors was 3 (range 3–4). The diagnosis of HCC was incidental in 36 (58 per cent) of the patients, mainly during follow-up of diabetes or investigation for abnormal liver function tests, as was found in more than 90 per cent of patients. Patients with normal and abnormal underlying parenchyma had similar preoperative characteristics including age, Model for End-stage Liver Disease (MELD) score, platelet count and aminotransferase levels (Table 1).

Table 1. Preoperative characteristics
 Overall (n = 62)Normal parenchyma (n = 24)Abnormal parenchyma (n = 38)P
  • Values in parentheses are percentages unless indicated otherwise;

  • *

    values are median (range). AST, aspartate aminotransferase; ALT, alanine aminotransferase; GGT, γ-glutamyltranspeptidase; MELD, Model for End-stage Liver Disease.

  • Fisher's exact test, except

  • Wilcoxon rank sum test (normal versus abnormal parenchyma).

Age (years)*70 (50–84)72 (53–81)68 (50–84)0·417
Body mass index (kg/m2)*30·4 (20·2–42·0)28·4 (23·2–36·2)31·1 (20·2–42·0)0·005
Hypertension56 (90)22 (92)34 (89)0·776
Diabetes52 (84)19 (79)33 (87)0·423
Dyslipidaemia40 (65)19 (79)21 (55)0·060
Incidental tumour36 (58)12 (50)24 (63)0·306
Prothrombin time (s)*98 (60–114)99 (69–114)97 (60–113)0·890
Platelets (× 109/l)*236 (86–474)272 (156–474)219 (86–442)0·450
AST (units/l)*36 (19–189)47 (19–189)36 (19–111)0·064
ALT (units/l)*43 (20–174)40 (21–159)46 (20–174)0·702
ALT/AST*1·1 (0·4–4·2)1·1 (0·4–1·9)1·2 (0·8–4·2)0·765
GGT (units/l)*94 (16–1566)106 (27–500)88 (16–1566)0·819
MELD score*7 (6–13)7 (6–13)7 (6–13)0·994
Normal α-fetoprotein level41 (66)14 (58)27 (71)0·205

Surgical procedure

Thirty-two patients (52 per cent) with MS-HCC underwent a major liver resection and 30 (48 per cent) had a minor resection. Nine of the patients having a major resection received preoperative PVE, which was associated with TACE in eight patients. Among the 38 patients with an abnormal parenchyma, those without severe fibrosis had preoperative PVE for anticipated major liver resection less frequently than those with severe fibrosis (3 of 12 versus 7 of 10 patients; P = 0·034). Extent of resection, blood loss, and rate and duration of hepatic inflow clamping were similar in patients with and without an abnormal parenchyma (Table 2).

Table 2. Operative characteristics
 Overall (n = 62)Normal parenchyma (n = 24)Abnormal parenchyma (n = 38)P
  • Values in parentheses are percentages unless indicated otherwise;

  • *

    values are median (range).

  • Fisher's exact test, except

  • Wilcoxon rank sum test (normal versus abnormal parenchyma).

Portal vein embolization9 (15)2 (8)7 (18)0·272
Major liver resection32 (52)15 (63)17 (45)0·173
Minor liver resection30 (48)9 (38)21 (55)0·173
Major right-sided resection22 (35)7 (29)15 (39)0·409
Associated procedure16 (26)8 (33)8 (21)0·210
Inflow clamping49 (79)20 (83)29 (76)0·509
Duration of clamping (min)*40 (3–120)35 (3–70)45 (8–120)0·990
Estimated blood loss (ml)*425 (50–2000)350 (50–1500)475 (100–2000)0·825
Transfusion14 (23)8 (33)6 (16)0·069
Packed red blood cells (units)*2·5 (1–5)2·0 (1–4)3·5 (1–5)0·420
Duration of surgery (min)*280 (75–570)330 (75–570)270 (140–440)0·027

Pathological analysis

Pathological examination of the underlying liver revealed that, of the 38 patients with an abnormal underlying parenchyma (61 per cent of the study population), 18 had severe fibrosis including 11 with cirrhosis (Table S1, supporting information). These patients had significantly lower rates of steatosis than patients without severe fibrosis (20 versus 45 per cent; P = 0·013). Of the 16 patients (26 per cent) who had NASH, eight showed no features of severe fibrosis. Among the 26 patients (42 per cent) who underwent preoperative liver biopsy of the non-tumorous parenchyma, there was total concordance between biopsy sample interpretation and final pathological assessment regarding steatosis, steatohepatitis and severe fibrosis for 17 specimens.

Tumour characteristics are shown in Table S2 (supporting information). One-third of the patients presented with lesions greater than 10 cm in size. Patients with a normal underlying liver had larger tumours (7·1 versus 9·4 cm; P = 0·012), which were more frequently associated with satellite nodules (13 of 24 versus 9 of 38; P = 0·014) and microvascular invasion (15 of 24 versus 14 of 38; P = 0·049). Seventeen patients (27 per cent) had lesions within the Milan criteria, including nine (15 per cent) with severe underlying fibrosis and only five (8 per cent) aged less than 65 years.

Postoperative outcome

Seven patients (11 per cent) died during the early postoperative period (Table S3, supporting information). All had an abnormal underlying parenchyma; five had no features of severe underlying fibrosis and five underwent major resection. Postoperative complications are detailed in Table 3. In the subgroup with an abnormal underlying liver, the mortality rate was not significantly different between patients with severe fibrosis and those with stage F0–F2 fibrosis and a NAS of at least 2 (2 of 18 versus 5 of 20 patients respectively; P = 0·270). Severe underlying fibrosis (P = 0·968), including stage F4 cirrhosis (P = 0·790), was not a risk factor for mortality.

Table 3. Postoperative complications
 Overall (n = 62)Normal parenchyma (n = 24)Abnormal parenchyma (n = 38)P
  • Values in parentheses are percentages unless indicated otherwise;

  • *

    values are median (range). ICU, intensive care unit.

  • Fisher's exact test, except

  • Wilcoxon rank sum test (normal versus abnormal parenchyma).

90-day mortality7 (11)0 (0)7 (18)0·026
Overall complications36 (58)11 (46)25 (66)0·120
Major complications19 (31)3 (13)16 (42)0·010
Liver-related major complications13 (21)1 (4)12 (32)0·014
Liver failure7 (11)0 (0)7 (18)0·026
Ascites14 (23)4 (17)10 (26)0·384
Biliary leakage3 (5)0 (0)3 (8)0·162
Haemorrhage3 (5)0 (0)3 (8)0·162
Cardiorespiratory major complications17 (27)3 (13)14 (37)0·036
Pulmonary embolism4 (6)1 (4)3 (8)0·567
Cardiac decompensation3 (5)0 (0)3 (8)0·162
Infectious complication19 (31)8 (33)11 (29)0·717
Duration of ICU stay (days)*4 (1–68)4 (1–68)5 (1–54)0·868
Duration of hospital stay (days)*12 (3–84)11 (3–84)13 (4–57)0·981

Major postoperative complications were observed in 19 patients (31 per cent) (Table 3) and occurred more frequently in patients with abnormal underlying liver (16 of 38 versus 3 of 24; P = 0·010), including patients with severe underlying fibrosis (7 of 18 versus 3 of 24 patients; P = 0·047) and those with stage F0–F2 fibrosis and a NAS of 2 or more (9 of 20 versus 3 of 24 patients; P = 0·016), compared with patients with normal underlying liver. Patients with stage F0–F2 and a NAS of 2 or above experienced major cardiorespiratory (9 of 20 versus 3 of 24 patients; P = 0·016) and liver-specific (5 of 20 versus 0 of 24 patients; P = 0·009) complications more frequently than those with normal underlying parenchyma. Multivariable analysis showed that age, inflow clamping, blood loss and the presence of stage F0–F2 fibrosis and a NAS of 2 or above in the underlying liver were associated with increased major complications (Table 4). Severe underlying fibrosis (P = 0·368), including F4 cirrhosis (P = 0·243), was not a risk factor for major complications.

Table 4. Univariable and multivariable analysis of predictive factors of major postoperative complications
 Major complications   
 Yes (n = 19)No (n = 43)PHazard ratioP
  • Values in parentheses are percentages unless indicated otherwise;

  • *

    values are median (range) and

  • 95 per cent confidence intervals. ALT, alanine aminotransferase; AST, aspartate aminotransferase; NAS, non-alcoholic fatty liver disease activity score; NASH, non-alcoholic steatohepatitis. The variables age above 70 years, transfusion, abnormal underlying liver and NASH were not analysed in the multivariable model because they were co-dependent with the variables age, blood loss, F0–F2 and NAS of 2 or above respectively.

  • Fisher's exact test, except

  • §

    Wilcoxon rank sum test.

Age (years)*73 (51–84)68 (50–81)0·063§1·11 (1·00, 1·23)0·043
Age > 70 years12 (63)15 (35)0·038  
Body mass index (kg/m2)*30·1 (20·2–40·4)30·4 (23·2–42·0)0·557§  
ALT/AST*1·2 (0·8–4·2)1·1 (0·4–2·4)0·118§  
Major resection12 (63)20 (47)0·227  
Major right-sided resection10 (53)12 (28)0·0542·02 (0·48, 8·41)0·342
Associated procedure6 (32)10 (23)0·490  
Clamping18 (95)31 (72)0·0431·47 (1·24, 1·81)0·021
Clamping time (min)*41·5 (8–120)40 (3–110)0·300§  
Blood loss (ml)*700 (100–2000)300 (100–1500)0·004§1·00 (1·00, 1·00)0·030
Transfusion7 (37)7 (16)0·074  
Tumour size (mm)*75 (35–220)78 (20–350)0·338§  
Abnormal underlying liver16 (84)22 (51)0·014  
 F0–F2 and NAS ≥ 29 (47)11 (26)0·0915·66 (1·21, 26·52)0·028
 F3–F47 (37)11 (26)0·368  
NASH8 (42)8 (19)0·051  

Survival

At a median follow-up of 24 months, 22 patients (35 per cent) were diagnosed with tumour recurrence; 11 of these patients died. One-year overall and disease-free survival rates were 83 and 83 per cent respectively, and 3-year rates were 75 and 70 per cent. There was a trend towards better long-term survival for patients with normal underlying liver compared with survival in those with abnormal underlying liver (90 versus 78 per cent at 1 year, and 90 versus 64 per cent at 3 years, respectively; P = 0·131) (Fig. S1, supporting information).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Disclosure
  8. References
  9. Supporting Information

The metabolic syndrome encompasses clinical and biological manifestations of insulin resistance. NAFLD, which defines hepatic manifestations of insulin resistance, ranges from simple steatosis to NASH1, 2 and can progress to cirrhosis. The metabolic syndrome is increasingly recognized as a co-risk factor for HCC20. In the present series of 62 patients with metabolic syndrome who underwent liver resection for HCC, only about 30 per cent of the patients developed HCC in underlying severely fibrotic livers. Thus, the present series confirms that HCC might arise in a context of insulin resistance even in the absence of severe underlying fibrosis1, 3–6, 21.

In the present study, the gradual increase over the years in the proportion of metabolic syndrome-associated HCCs, amongst other aetiologies, probably accounts for both better classification and rising incidence of this disease. Almost 40 per cent of the patients presented with minimal parenchymal lesions and would probably have been classified as having ‘HCC developed on normal liver’, even as recently as 10 years ago. The rising incidence of the metabolic syndrome in the general population—up to 24 per cent in the USA, especially in elderly patients12—suggests that surgeons can expect increasing numbers of patients to be referred to hepatopancreatobiliary units for HCC in this context.

As for other causes of HCC22, it is likely that patients with metabolic syndrome may benefit from earlier diagnosis of the tumour. In fact, HCC was diagnosed incidentally in nearly 60 per cent of patients, mostly at an advanced stage, and with more than 90 per cent of patients showing abnormal liver function. This suggests that liver imaging in the presence of unusual liver test results could aid earlier tumour diagnosis in patients with metabolic syndrome.

Over recent years, continuing refinements in surgical technique, patient selection and perioperative care have led to increasingly safer liver resection23. However, these advances have not translated wholly into improved care for the specific population of patients with MS-HCC and underlying liver disease. In this subgroup, the high mortality rate observed in the present study might be explained by the underestimation of patients' co-morbidity. Advanced age, obesity and diabetes, which are all known to affect the postoperative course negatively24, were found in more than 80 per cent of patients. Furthermore, the observed concordance between the higher incidence of major postoperative cardiorespiratory events and the presence of a fatty underlying liver confirms the independent association of NAFLD, an intense proinflammatory state secondary to advanced insulin resistance, and increased vulnerability following impaired cardiorespiratory status25. This finding suggests that specific preoperative cardiorespiratory and anaesthetic evaluation should probably be undertaken in these patients.

The most striking finding in this series was the relationship between the operative risk and the presence of an abnormal underlying parenchyma, especially in patients without severe fibrosis. Similar to major hepatectomy for colorectal liver metastases in patients after prolonged chemotherapy26, and in line with other studies27–29, the present results underline the influence of significant steatosis on operative outcome. Moreover, the increased morbidity and mortality rates observed in patients with a NAS of 2 or more suggest that the operative risk is further enhanced in the presence of associated inflammatory processes.

Based on these results, it would seem reasonable to suggest that improvements in perioperative management should be undertaken, especially for patients with stage F0–F2 fibrosis and a NAS of 2 or above. In fact, when major resection was planned, these patients had preoperative PVE less frequently than those with stage F3–F4. This may indicate that there was an underestimation of the operative risk, especially for patients without severe underlying fibrosis, and therefore inadequate preoperative and operative management. It is thus likely that these patients would benefit from preoperative assessment of the underlying parenchyma using percutaneous biopsy or magnetic resonance imaging30 and parenchyma-sparing resection31, as well as careful use of hepatic inflow clamping, as in patients with cirrhosis32.

The final interesting result of this study was the good long-term prognosis of patients with MS-HCC. Indeed, both 1- and 3-year overall and disease-free survival rates (83 and 75 per cent, and 83 and 70 per cent, respectively) compare favourably with data reported recently on resection of HCC in chronic liver disease of other aetiology5. Interestingly, patients with normal underlying parenchyma had more advanced disease, including larger tumours, more frequent satellite nodules and higher microvascular invasion rates. This is most likely a consequence of the delayed diagnosis in the absence of specific symptoms. Accordingly, the trend towards better survival observed in patients with normal underlying liver cannot be attributed to more favourable biology, but should rather be related to increased mortality rates in patients with abnormal underlying liver. Nevertheless, the good long-term results in this group of patients support a radical treatment, provided that surgeons are aware of the particular hazards of liver resection and thus offer optimal perioperative care to these patients.

The present study has several limitations owing to both its retrospective nature and the small number of patients included over a long period. The retrieved factors for the metabolic syndrome did not all fulfil the consensus definition19, and the assumption that all patients taking a statin or fenofibrate had dyslipidaemia might have led to the inclusion of some patients who were receiving such medication for primary cardiovascular prevention. Similarly, given the higher enrolment of patients with metabolic syndrome during the later years of the study, it is likely that earlier in the study some patients with metabolic syndrome with incomplete diagnostic workup may have been misclassified. Finally, considering the high mortality rate, one could legitimately wonder whether liver transplantation might be the appropriate treatment for these patients5, 33. However, as only a small number of patients had lesions within the Milan criteria, it appears that liver resection is presently the only curative option for the majority of these patients.

Disclosure

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Disclosure
  8. References
  9. Supporting Information

The authors declare no conflict of interest.

Supporting information

Additional supporting information may be found in the online version of this article:

Table S1 Pathological analysis of the underlying liver (Word document)

Table S2 Tumour characteristics (Word document)

Table S3 Characteristics of patients who died (Word document)

Fig. S1 Overall survival according to the underlying liver (Word document)

Please note: John Wiley & Sons Ltd is not responsible for the functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Disclosure
  8. References
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Disclosure
  8. References
  9. Supporting Information
FilenameFormatSizeDescription
bjs_8963_sm_suppinfo.doc103KSupporting Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.