444 Original article
Major hepatectomy is safe for hepatocellular carcinoma in elderly patients with cirrhosis Wan-li Wanga,c, Ying Zhua, Ji-wen Chenga, Mu-xing Lia, Jian-min Xiaa, Jie Haoa, Liang Yua, Yi Lva,b, Zheng Wua and Bo Wanga Background Because of an increasing aging population worldwide, a greater number of elderly patients are being considered for hepatic resection. The objective of this retrospective pair-matched study was to assess the influence of age on postoperative outcomes after major hepatectomy (resection of three or more Couinaud segments) in elderly patients with hepatocellular carcinoma (HCC) and cirrhosis. Patients and methods A retrospective review of patient demographics, diagnoses, surgical treatments, and early postoperative outcomes was performed. Results A total of 208 HCC patients with cirrhosis underwent major hepatectomy between 2007 and 2012. The mortality rate was 3.57% in patients aged 70 years or more (group E) compared with 1.32% in those aged below 70 years (group Y; P = 0.630). The overall complication rates were 53.57% in group E and 47.37% in group Y (P = 0.427). Increasing age was independently associated with postoperative pneumonia (P < 0.001), bacteremia (P = 0.026), and respiratory failure requiring reintubation (P = 0.028). A total of 25.00% of patients had a Clavien–Dindo classification grade of 3a or more in group E compared with 13.16% in group Y (P = 0.040). In multivariate analysis, intraoperative red blood cell transfusion of 5 U or more (P = 0.016; hazard ratio 4.812;
Introduction Hepatocellular carcinoma (HCC) is one of the most frequent neoplasms worldwide. In the majority of cases, particularly in Asia, it is associated with concomitant liver cirrhosis due to chronic hepatitis virus infection [1–3], which may limit the possibility of surgical treatment. Liver resection is regarded as the most potentially curative therapy for HCC patients when liver transplantation is not an optimum management option or immediately accessible [4,5]. However, liver resection is still a complex procedure associated with high morbidity and mortality rates, especially in patients with cirrhosis [6,7]. The situation is even more serious for elderly patients because of a reduced life expectancy compared with younger patients [1]. Owing to improvements in surgical techniques [8,9], anesthesia and medical care, and better perioperative assessments [10], the rates of morbidity and mortality after liver resection have notably decreased [11–13]. However, liver resection in elderly patients still carries a high risk, especially for those with c 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins 0954-691X
95% confidence interval 1.332–17.384) was a predictor of higher morbidity in the elderly. Conclusion With rigorous screening of patients and improvement of perioperative management and operative techniques, major hepatectomy can be safely performed on HCC patients aged 70 years or more with liver cirrhosis. Intraoperative red blood cell transfusion of 5 U or more was predictive of higher morbidity in the elderly. Surgeons should take care to minimize the likelihood of intraoperative blood transfusion in elderly patients. Eur J Gastroenterol Hepatol c 2014 Wolters Kluwer Health | Lippincott 26:444–451 Williams & Wilkins. European Journal of Gastroenterology & Hepatology 2014, 26:444–451 Keywords: cirrhosis, complications, elderly, hepatocellular carcinoma, major hepatectomy a Department of Hepatobiliary Surgery, First Affiliated Hospital, School of Medicine, bInstitute of Advanced Surgical Technology and Engineering, Xi’an Jiaotong University, Xi’an and cDepartment of General Surgery, Bazhong Central Hospital, Bazhong, People’s Republic of China
Correspondence to Bo Wang, MD, PhD, Department of Hepatobiliary Surgery, First Affiliated Hospital, School of Medicine, Xi’an Jiaotong University, No. 277 West Yanta Road, Xi’an 710061, Shaanxi, People’s Republic of China Tel/fax: + 86 029 8532 3902; e-mail: [email protected] Received 5 November 2013 Accepted 17 December 2013
concomitant liver cirrhosis. Age-associated changes in liver function may be caused by several mechanisms, including: (a) a decline in liver volume, (b) an increase in the hepatic dense body compartment (lipofuscin), (c) moderate declines in the phase I metabolism of certain drugs, (d) shifts in the expression of a variety of proteins, (e) diminished hepatopancreaticobiliary (HPB) function, and (f) other more subtle changes (e.g. muted responses to oxidative stress, reduced expression of regulatory growth genes, diminished rates of DNA repair, and telomere shortening) [14]. These changes may contribute to the reduced hepatic regenerative capacity in the elderly [14], and this may compound the risk associated with major hepatectomy and advanced age. It has been reported that liver resections for HCC and colorectal liver metastases are as safe in the elderly as those in younger patients [5,15,16]. However, few studies have focused on the outcomes after major hepatectomy (resection of three or more Couinaud segments) in patients with liver cirrhosis aged over 70 years. DOI: 10.1097/MEG.0000000000000046
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Major hepatectomy for HCC in elderly patients Wang et al.
In this study, the outcome of patients with HCC and concomitant liver cirrhosis who underwent major hepatectomy was studied retrospectively, and the influence of age on early postoperative recovery was evaluated. The aim was to compare the rates and severity of postoperative complications following major liver resection in two different age groups.
Patients and methods This retrospective pair-matched study was carried out after obtaining institutional review board approval from the Medical College of Xi’an Jiaotong University. Written consent was given by patients for their information to be stored in the hospital database and used for research. All treatment strategies and indications for surgery were discussed by a multidisciplinary board. Before surgery, all patients had been informed of the necessity of surgery and the potential for perioperative complications, they all agreed to surgery and gave informed consent. Patients aged 70 years or more (group E) who underwent major hepatectomy between 2007 and 2012 were matched with those aged less than 70 years (group Y) who underwent major hepatectomy, according to diagnosis and the extent of liver resection. The study included only HCC patients with concomitant liver cirrhosis who underwent major hepatectomy. Patients younger than 18 years, those with noncirrhotic HCC, and those who had a minor resection were excluded from the analysis. Postoperative morbidity and mortality was defined by the Clavien–Dindo classification system [17] and liver failure (LF) by the ‘50–50’ criteria at postoperative day 5 [18]. Postoperative mortality included death within 90 days of liver resection. The Clavien–Dindo classification of surgical complications within 30 days of the postoperative period was also used to grade the severity of postoperative complications. Postoperative hemorrhage was considered when bleeding from the drainage tubes exceeded 100 ml/h. The diagnostic criteria of pneumonia are based on imaging and sputum culture, and bacteremia are based on blood culture. Patient demographics, diagnoses, preoperative condition, surgical procedures, and postoperative outcomes were systematically recorded in different groups, and the data were analyzed retrospectively. Preoperative, intraoperative, and postoperative data were collected prospectively in a computerized database. The data were imported into SPSS software (version 16.0; SPSS Inc., Chicago, Illinois, USA) for analysis. Univariate w2 and Fisher’s exact test (in cases of low frequency) were performed to discriminate discrete variables. The data were non-normally distributed, and therefore nonparametric statistical analyses were used. Mann–Whitney U-tests were performed to test the differences between groups with continuous variables, and w2 analysis was
445
performed for categorical variables. Significance was set at P value less than 0.05. Univariate analysis and multivariate analysis were performed to determine possible predictors of complications in elderly patients.
Perioperative management
Before surgery, a senior HPB surgeon and an anesthetist assessed all patients. Preoperative investigation of the patients included blood biochemistry, a-fetoprotein assay, chest radiograph, ECG, percutaneous ultrasonography, high-resolution contrast-enhanced computed tomography scan of the chest and abdomen, gastroscope, and/or indocyanine green clearance test. The aim of imaging was to assess the liver volume and future remnant liver volume, vascular anatomy, portal hypertension, the extent of the hepatic lesion and to rule out extrahepatic malignant disease. The diagnosis of cirrhosis was made according to the guidelines published by Desmet et al. [19]. All cases were pathologically confirmed HCC with concomitant cirrhosis. All patients’ future remnant liver volume was more than 40%. Liver function was assessed by Child–Pugh classification and/or indocyanine green retention test. All patients older than 65 years were assessed for cardiac risk by a dedicated preoperative workup according to the guidelines of European Society of Anesthesiology (ESA) [20]. Liver resections were classified according to the Brisbane nomenclature [21]. Intermittent vascular occlusion was performed with cycles of 15-min inflow occlusion followed by 5 min of reperfusion. Central venous pressure was kept below 5 mmHg during liver parenchymal transection. Large vessels and bile ducts were ligatured with a suture, clip or vascular stapling device or secured intrahepatically with interrupted sutures. Close-suction drainage tube was selectively placed in the subphrenic or subhepatic space close to the transection surface at the end of the operation according to intraoperative conditions, and was unplugged at the most opportune time. All patients were nursed in the ICU during the early postoperative period to maintain the fluid balance, and for cardiac monitoring and respiratory management. Postoperative liver and kidney function and routine blood tests were assessed. On the basis of the liver function tests, some patients were treated with albumin or fresh frozen plasma to maintain postoperative albumin levels above 35 g/l to avoid ascites or pleural effusion. Early oral feeding was encouraged and parenteral nutrition was started immediately after major hepatectomy for all patients. Careful attention was paid to the management of intravenous fluid and electrolyte balance during the early postoperative period. Intravenous fluid was rigidly restricted to less than 2 l/day to minimize sodium and fluid retention and reduce the potential for ascites accumulation in all patients. Transcutaneous ultrasonography was performed on patients on postoperative day 3 or 7 by
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
446 European Journal of Gastroenterology & Hepatology 2014, Vol 26 No 4
Table 1 Preoperative characteristics of patients undergoing major hepatectomy for hepatocellular carcinoma with concomitant liver cirrhosis Variables Median age (range) (years) Sex (male/female) HBV-related cirrhosis [n (%)] HCV-related cirrhosis [n (%)] Diameter of tumor (cm) Number of tumor History of abdominal surgery [n (%)] History of alimentary tract hemorrhage [n (%)] History of TACE [n (%)] Smoking [n (%)] Comorbidities Hypertension Diabetes Cardiac arrhythmias COPD Other Platelet count ( 109/l) Hemoglobin (g/l) White blood cell count ( 109/l) Prothrombin time (s) ALT (IU/l) AST (IU/l) Total bilirubin (mmol/l) Total cholesterol (mmol/l) Albumin (g/l) Serum creatinine Serum sodium Serum AFP (ng/ml) Child–Pugh grade (n) Class A Class B ASA grade (n) Grade 1 Grade 2 Grade 3
Group E (Z 70 years; n = 56) 74 (70–78) 40/16 54 (96.42) 2 (3.58) 7.50 (3.00–15.00) 1.18±0.61 14 (25.00) 8 (14.29) 8 (14.29) 16 (28.57)
Group Y (< 70 years; n = 152) 48 (19–69) 120/32 150 (98.68) 2 (1.32) 8.85 (2.40–17.50) 1.20±0.56 24 (15.79) 12 (7.89) 16 (10.53) 52 (34.21)
20 4 6 4 0 119.00 (36.00–500.00) 124.46±18.48 4.71 (1.80–70.00) 13.02±1.46 35.50 (11.00–184.00) 40.50 (17.00–205.00) 16.95 (4.62–338.90) 3.86±1.04 38.62 (29.09–47.65) 74.66 (62.58–105.79) 140.72 (131.62–144.70) 97.02 (2.47–31104.00)
16 4 6 8 3 135.50 (17.00–348.00) 127.32±16.55 5.12 (1.40–20.29) 13.36±1.37 45.50 (13.00–414.00) 52.00 (12.00–619.00) 15.77 (5.05–378.92) 3.76±1.22 38.72 (15.62–47.60) 75.26 (44.69–117.70) 139.52 (132.00–146.50) 1210 (1.89–121000.00)
48 8
122 30
0 38 18
10 92 50
P value < 0.001 0.254 0.630 0.630 0.122 0.014 0.127 0.165 0.452 0.442 < 0.001 0.274 0.128 0.857 0.687 0.946 < 0.001 0.759 < 0.001 0.043 0.035 0.714 < 0.001 0.626 0.159 0.611 0.107 0.367
0.113
AFP, a-fetoprotein; ALT, alanine aminotransferase; AST, aspartate aminotransferase; ASA, American Society of Anesthesiologists; COPD, chronic obstructive pulmonary disease; HBV, hepatitis B virus; HCV, hepatitis C virus; TACE, transcatheter arterial chemoembolization.
experienced radiologists to detect any perihepatic or celiac collection of ascitic fluid or pleural effusion.
Results Demographics and perioperative details
There were 1018 consecutive liver resections performed between 2007 and 2012 at the Department of Hepatobiliary Surgery, First Affiliated Hospital of Medical College, Xi’an Jiaotong University, Xi’an, China. Among these patients, a total of 208 histologically confirmed HCC patients with concomitant cirrhosis underwent major hepatectomy. We identified 56 patients aged 70 years or more with major liver resection who could be matched with 152 control patients aged below 70 years undergoing similar types of liver resection. Demographics, preoperative comorbidities, and preoperative laboratory values are shown in Table 1. The male-tofemale ratio was 40 : 16 and 120 : 32 in groups E and Y, respectively (P = 0.254). The median age was 74 years (range 70–78 years) and 48 years (range 19–69 years) in groups E and Y, respectively. Most patients had hepatitis B virus (HBV)-related cirrhosis (96.42% in group E vs. 98.68% in group Y; P = 0.630). A higher rate of hypertension was reported in group E (P < 0.001). There were also
significant differences in terms of the number of tumors (P = 0.014), preoperative hemoglobin (P < 0.001), prothrombin time (P < 0.001), and preoperative total cholesterol (P < 0.001), favoring group Y. However, there were significant differences in terms of alanine aminotransferase (ALT) (P = 0.043) and aspartate aminotransferase (AST) (P = 0.035), favoring group E. There were no differences in terms of other clinical variables. Intraoperative and postoperative results for both groups are shown in Table 2; there were no significant differences between the two groups. Postoperative changes in serum ALT, AST, total bilirubin (TBIL) and albumin (ALB) in the two groups are shown in Fig. 1. The changes in ALT, AST, and ALB were not significantly different between the two groups (P = 0.903, P = 0.791, and P = 0.594, respectively). However, there were significant differences in TBIL between the two groups (P = 0.026). There was no difference in postoperative albumin infusion between the two groups (P = 0.353). Postoperative complications
Postoperative complications in both groups are shown in Table 3. The mortality rate was 3.57% (two deaths) in
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Major hepatectomy for HCC in elderly patients Wang et al. 447
Table 2
Risk factor analysis
Intraoperative or postoperative results
Variables
Group E (Z 70 years; n = 56)
Group Y (< 70 years; n = 152)
Type of resection [n (%)] Right hepatectomy 24 (42.86) 44 (28.95) Left hepatectomy 10 (17.86) 38 (25.00) Extended right 0 (0.00) 2 (1.32) hepatectomy Extended left hepatectomy 0 (0.00) 6 (3.95) Resection of three 22 (39.29) 62 (40.79) segments Simultaneous extrahepatic procedures [n (%)] Diaphragm resection/repair 6 (10.71) 4 (2.63) Bile duct excision 4 (7.14) 12 (7.89) Cholangiojejunostomy 6 (10.71) 4 (2.63) Other upper 2 (3.57) 2 (1.32) gastrointestinal resection Portal vein tumor thrombus 14 (25.00) 22 (14.47) Splenic artery ligation 2 (3.57) 4 (2.63) Portal vein reconstruct 16 (28.57) 26 (17.11) Other 2 (3.57) 6 (3.95) Simultaneous ablation 8 (14.29) 12 (7.89) Intraoperative blood loss (ml) 950 (200–4000) 850 (100–7500) Intraoperative RBC 4 (2–18) 4 (0–24) transfusion (U) Hepatic portal occlusion (min) 14.0 (0.00–54.00) 15.50 (0.00–58.00) Operative time (min) 245 (140–490) 280 (90–480) Postoperative albumin 90 (0–450) 75 (0–380) transfusion (g) ICU stay (days) 3 (2–6) 3 (1–11) Postoperative hospital stay 15 (6–32) 15 (5–42) (days)
P value 0.225
0.563
On univariate analysis, intraoperative blood loss of 1000 ml or more (P = 0.021), intraoperative red blood cell (RBC) transfusion of 5 U or more (P = 0.038), and operative time of 250 min or more (P = 0.027) were found to be predictive of postoperative complications (Table 4). In multivariate analysis, intraoperative RBC transfusion of 5 U or more (P = 0.016; hazard ratio 4.812; 95% CI 1.332–17.384) was a predictor of higher morbidity in the elderly (Table 5). In addition, patients with complications had a longer stay in the ICU (P = 0.015) and in hospital (P = 0.043) than those without complications.
Discussion
0.165 0.713 0.728 0.376 0.113 0.353 0.480 0.975
RBC, red blood cell.
group E compared with 1.32% (two deaths) in group Y, and this was not significantly different [P = 0.630; odds ratio (OR) 2.778; 95% confidence interval (CI) 0.382–20.209]. In both groups, the causes of death were LF. Overall complication rates were 53.57% in group E and 47.37% in group Y; this difference was not significant (P = 0.427; OR 1.282; 95% CI 0.694–2.369). A total of 25.00% of patients experienced complications requiring invasive interventions (Clavien–Dindo classification grade Z 3a) in group E compared with 13.16% in group Y (P = 0.040; OR 2.200; 95% CI 1.023–4.733). A number of postoperative complications were more common in group E including: postoperative pneumonia (P < 0.001), bacteremia (P = 0.026), and respiratory failure requiring reintubation (P = 0.028). Two elderly patients had postoperative confusion related to liver insufficiency, and they recovered with conservative treatment. With regard to liver-related complications, there was one bile leak in an elderly patient, which required percutaneous drainage. There were four cases of bile leak in group Y, two received endoscopic nasobiliary drainage, one required percutaneous drainage, and one required percutaneous drainage of biloma. None of the patients died due to bile leakage. In addition, LF occurred in four patients in group E and six in group Y, and there was no significant difference (P = 0.555).
An aging worldwide population has resulted in a significant increase in the number of elderly patients considered for resection of malignant HPB tumors. Malignant HPB tumors most commonly occur during the sixth to eighth decades of life [22]. In addition, the majority of patients with HCC have underlying cirrhosis [11,12]. Previous studies have showed that liver resection is associated with high morbidity and mortality rates, especially in patients with cirrhosis [23–25]. With the current improvements in surgical techniques and perioperative care, major hepatic resection for HCC, mainly in the form of classical right or left hepatectomy, is associated with acceptable operative morbidity and mortality in selected patients with cirrhosis [11,26,27]. Some studies also demonstrate that major hepatectomy can be performed as safely in elderly patients as in younger patients [5,15,16,28]. However, there is a paucity of data in the literature on major resection for HCC in cirrhotic elderly patients. There are a number of biases against major hepatectomy for elderly HCC patients with concomitant liver cirrhosis including: (i) elderly patients have less aggressive tumors compared with their younger counterparts [29]; (ii) an increased incidence of medical comorbidities leads to a reduced tolerance to surgery [30]; and (iii) aging leads to a number of structural and functional changes in the liver that may make liver resection less tolerable [14]. Furthermore, despite recent advances in surgical techniques and perioperative management, hepatic resection in the cirrhotic liver remains a major challenge for surgeons because of the specific risks associated with cirrhosis, such as difficult mobilization and transection of the fibrotic liver, increased bleeding as a result of portal hypertension, and higher risk of LF [30]. Resection of HCC in cirrhotic liver may be fruitless as there is a high risk of recurrence in the remnant cirrhotic liver [4,23,24]. Local therapy for small HCC, such as percutaneous alcohol injection, transcatheter hepatic arterial chemoembolization, and radiofrequency ablation, gives long-term results similar to those of surgery [31]. Some studies have revealed that elderly HCC patients with cirrhosis after major liver resection derive as much
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
448 European Journal of Gastroenterology & Hepatology 2014, Vol 26 No 4
Fig. 1
(a)
(b) 600.00
P = 0.903
P = 0.791 600.00 AST in U/l (median)
ALT in U/l (median)
500.00 400.00 300.00 200.00
400.00
200.00
100.00 0.00
0.00 PO
POD1
POD3
POD7
(c) P = 0.026
60.00
POD1
POD3
POD7
50.00 40.00 30.00
P = 0.594
45.00
ALB in g/l (median)
TBIL in μmol/l (median)
PO (d)
20.00 10.00
40.00
35.00
30.00
25.00 PO
POD1
POD3
POD7 Group E
PO
POD1
POD3
POD7
Group Y
Serum levels of alanine aminotransferase (ALT), aspartate aminotransferase (AST), total bilirubin (TBIL), and albumin (ALB). (a) Median perioperative changes in levels of ALT in U/l, (b) AST in U/l, (c) TBIL in mmol/l, and (d) ALB in g/l. The smallest and the largest observations are represented by the ends of the error bars (5th percentile and 95th percentile, respectively). The horizontal solid lines represent the normal values for each parameter. There was no significant difference in preoperative and postoperative serum levels of ALT, AST, and ALB between the groups. There was a significant difference in preoperative and postoperative serum levels of TBIL between the groups (P = 0.026). PO, preoperation; POD, postoperative day.
benefit as their younger counterparts [12,27,32]. In the current study, the 53.57% morbidity and 3.57% mortality rates in elderly HCC patients with cirrhosis are comparable with the rates reported in the published literature [4,11,27]. Our study demonstrates that major liver resection can be performed in HCC patients with cirrhosis aged 70 years or more with comparable early outcomes to those observed in younger patients. Advanced age is not the major determinant of the morbidity and mortality rates associated with major liver resection in elderly HCC patients with cirrhosis. This provides important evidence to dispel the common misconception that all elderly HCC patients with cirrhosis belong to a high surgical risk group by dint of their advanced age alone. Therefore, each elderly HCC patient with cirrhosis should be assessed for individual risk, and advanced age
should not be considered a contraindication for major liver resection. It is well established that the function of most organs usually deteriorates with age. In this study, the incidence of hypertension, and preoperative total cholesterol levels were significantly higher, and hemoglobin level was significantly lower in group E compared with group Y. This may reflect a decreased liver reserve capacity in the elderly. However, these factors did not result in an increase in mortality and morbidity rates in the elderly. Preoperative ALT and AST were significantly higher in group Y than in group E. This may reflect the aggressive selection of patients in group Y and prudent selection of patients in group E. Furthermore, because there was no significant difference between the two groups in the American Society of Anesthesiologists
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Major hepatectomy for HCC in elderly patients Wang et al. 449
Table 3
Details of postoperative complications
Complications Postoperative morbidity [n (%)] Postoperative mortality [n (%)] Confusion Wound infection Pneumonia Pleural effusion Pleural effusion requiring drainage Ascites Ascites requiring drainage Bile leak Hemorrhage Bacteremia Arrhythmia Liver failure Other organ failure Reoperation Respiratory failure requiring reintubation MODS Clavien–Dindo classification grade of complications No complication Grade 1 Grade 2 Grade 3a Grade 3b Grade 4a Grade 4b Grade 5
Group E (Z 70 years; n = 56)
Group Y (< 70 years; n = 152)
P value
30 (53.57) 2 (3.57) 2 4 12 8 4 24 18 1 4 3 6 4 2 0 4
72 (47.37) 2 (1.32) 0 4 6 16 6 48 36 4 12 0 5 6 1 1 1
0.427 0.630 0.123 0.274 < 0.001 0.452 0.555 0.129 0.217 0.999 0.999 0.026 0.076 0.555 0.364 0.999 0.028
4
2
26 4 22 6 0 6 2 2
80 14 64 12 2 6 2 2
0.078 0.389
MODS, multiple organ dysfunction syndrome.
(ASA) grade, the severity of associated diseases may be comparable between the groups. Although the recovery of postoperative serum TBIL level was inferior in elderly patients, they did not have higher liver insufficiencyrelated morbidity than younger patients. Elderly patients had similar changes in postoperative ALT, AST, and ALB values when compared with the younger patients. In view of these findings, we could not arbitrarily find any evidence that elderly patients were more susceptible to liver insufficiency than younger patients. Our study is noteworthy in that we have recorded all complications in the postoperative period using the Clavien– Dindo complication classification grade system, whereas many of the published studies have only reported complication rates [4,11,27]. For those studies that confuse the types of complications, treatment interventions and therefore the severity of complications were not uniformly disclosed, suggesting our complication rate may be higher than other reports. Of note, older age was associated with more severe complications (Z grade 3a) requiring invasive procedures. Therefore, this suggests that while elderly patients require more intense monitoring and greater nursing care and physical therapy, they can expect similar surgical outcomes. In a multivariate analysis, intraoperative RBC transfusion of 5 U or more (P = 0.016; hazard ratio 4.812; 95% CI 1.332–17.384) was a predictor of higher morbidity in the elderly. Bernard et al. [33] concluded that intraoperative
transfusion of 1–2 U packed RBCs is associated with increased 30-day mortality, surgical-site infection, pneumonia, and sepsis in general surgery patients. Therefore, intraoperative blood transfusion must strictly abide by transfusion guidelines. Interestingly, however, time of hepatic portal occlusion, simultaneous extrahepatic procedures and number of Couinaud liver segments resected did not correlate with complication rates, suggesting that the complexity of surgery itself was not responsible for the higher complication rates in the elderly. Overall, 75–80% of HCC is attributable to chronic HBV (50–55%) or hepatitis C virus (25–30%) infection [34–37]. Furthermore, HBV infection may be higher in Asia [11]. In our study, all patients had HCC with cirrhosis, indicating that HBV infection was more common. These patients had better liver function and fewer tumors than younger patients. Therefore, these patients may be good candidates for major liver resection. It should be noted, however, that numerous studies have reported that resection for HCC can reactive HBV, initiate virus replication, and even result in LF or cancer recurrence. Therefore, antiviral therapy seems to be an appropriate option, and should even be a part of the protocol for all HBV-related HCC patients who undergo liver resection, irrespective of whether the viral loads are high or low [38,39]. Unfortunately, in the current study there is a lack of data on antiviral treatment and patients were not followed for longer observation periods. Therefore, we could not determine the benefit of antiviral treatment in elderly HBV-related HCC patients with cirrhosis. This study shows that major hepatectomy can be performed in HCC patients aged 70 years or more with concomitant liver cirrhosis with comparable outcomes to those observed in younger patients. However, our study is limited by the small sample size, a retrospective analysis, rigorous selection criteria, and a lack of antiviral treatment data. The statistical power may be insufficient to detect a small susceptible effect of some factors on univariate and multivariate analysis. Therefore, prospective studies in larger samples of patients with aggressive surgical procedures (such as extended right/left hepatectomy) and longer observation periods are needed to clarify the effects of age on elderly HCC patients with cirrhosis. In conclusion, major liver resection can be relatively safely performed in carefully selected elderly HCC patients with concomitant liver cirrhosis. Advanced age alone should not be considered a contraindication for major liver resection. With improving perioperative assessment and operative techniques, most complications after cirrhotic liver resection can be treated and have a low mortality rate. However, more care should be taken if intraoperative RBC transfusion of more than 5 U is required. Surgeons should take care to minimize the likelihood of intraoperative blood transfusion in elderly patients.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
450
European Journal of Gastroenterology & Hepatology 2014, Vol 26 No 4
Table 4
Univariate analysis of variables predictive of complications in elderly patients
Covariables
Affected by complications (%)
Sex Male Female Age History of abdominal surgery History of alimentary tract hemorrhage History of TACE Smoking Hypertension Diabetes Cardiac arrhythmias COPD Platelet count (109/l) White blood cell count ( 109/l) Hemoglobin (g/l) Prothrombin time (s) Preoperative ALT (IU/l) Preoperative AST (IU/l) Preoperative total bilirubin (mmol/l) Total cholesterol (mmol/l) Serum creatinine Preoperative albumin (g/l) Serum sodium Serum AFP (ng/ml) > 400 (%) Child–Pugh grade Class A Class B ASA grade (%) Grade 1 Grade 2 Grade 3 Type of resection (%) Right hepatectomy Left hepatectomy Number of liver segment resected Z4 3 Simultaneous extrahepatic procedures Simultaneous ablation Intraoperative blood loss (ml) Z 1000 < 1000 Intraoperative RBC transfusion (U) Z5
Major hepatectomy is safe for hepatocellular carcinoma in elderly patients with cirrhosis Wan-li Wanga,c, Ying Zhua, Ji-wen Chenga, Mu-xing Lia, Jian-min Xiaa, Jie Haoa, Liang Yua, Yi Lva,b, Zheng Wua and Bo Wanga Background Because of an increasing aging population worldwide, a greater number of elderly patients are being considered for hepatic resection. The objective of this retrospective pair-matched study was to assess the influence of age on postoperative outcomes after major hepatectomy (resection of three or more Couinaud segments) in elderly patients with hepatocellular carcinoma (HCC) and cirrhosis. Patients and methods A retrospective review of patient demographics, diagnoses, surgical treatments, and early postoperative outcomes was performed. Results A total of 208 HCC patients with cirrhosis underwent major hepatectomy between 2007 and 2012. The mortality rate was 3.57% in patients aged 70 years or more (group E) compared with 1.32% in those aged below 70 years (group Y; P = 0.630). The overall complication rates were 53.57% in group E and 47.37% in group Y (P = 0.427). Increasing age was independently associated with postoperative pneumonia (P < 0.001), bacteremia (P = 0.026), and respiratory failure requiring reintubation (P = 0.028). A total of 25.00% of patients had a Clavien–Dindo classification grade of 3a or more in group E compared with 13.16% in group Y (P = 0.040). In multivariate analysis, intraoperative red blood cell transfusion of 5 U or more (P = 0.016; hazard ratio 4.812;
Introduction Hepatocellular carcinoma (HCC) is one of the most frequent neoplasms worldwide. In the majority of cases, particularly in Asia, it is associated with concomitant liver cirrhosis due to chronic hepatitis virus infection [1–3], which may limit the possibility of surgical treatment. Liver resection is regarded as the most potentially curative therapy for HCC patients when liver transplantation is not an optimum management option or immediately accessible [4,5]. However, liver resection is still a complex procedure associated with high morbidity and mortality rates, especially in patients with cirrhosis [6,7]. The situation is even more serious for elderly patients because of a reduced life expectancy compared with younger patients [1]. Owing to improvements in surgical techniques [8,9], anesthesia and medical care, and better perioperative assessments [10], the rates of morbidity and mortality after liver resection have notably decreased [11–13]. However, liver resection in elderly patients still carries a high risk, especially for those with c 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins 0954-691X
95% confidence interval 1.332–17.384) was a predictor of higher morbidity in the elderly. Conclusion With rigorous screening of patients and improvement of perioperative management and operative techniques, major hepatectomy can be safely performed on HCC patients aged 70 years or more with liver cirrhosis. Intraoperative red blood cell transfusion of 5 U or more was predictive of higher morbidity in the elderly. Surgeons should take care to minimize the likelihood of intraoperative blood transfusion in elderly patients. Eur J Gastroenterol Hepatol c 2014 Wolters Kluwer Health | Lippincott 26:444–451 Williams & Wilkins. European Journal of Gastroenterology & Hepatology 2014, 26:444–451 Keywords: cirrhosis, complications, elderly, hepatocellular carcinoma, major hepatectomy a Department of Hepatobiliary Surgery, First Affiliated Hospital, School of Medicine, bInstitute of Advanced Surgical Technology and Engineering, Xi’an Jiaotong University, Xi’an and cDepartment of General Surgery, Bazhong Central Hospital, Bazhong, People’s Republic of China
Correspondence to Bo Wang, MD, PhD, Department of Hepatobiliary Surgery, First Affiliated Hospital, School of Medicine, Xi’an Jiaotong University, No. 277 West Yanta Road, Xi’an 710061, Shaanxi, People’s Republic of China Tel/fax: + 86 029 8532 3902; e-mail: [email protected] Received 5 November 2013 Accepted 17 December 2013
concomitant liver cirrhosis. Age-associated changes in liver function may be caused by several mechanisms, including: (a) a decline in liver volume, (b) an increase in the hepatic dense body compartment (lipofuscin), (c) moderate declines in the phase I metabolism of certain drugs, (d) shifts in the expression of a variety of proteins, (e) diminished hepatopancreaticobiliary (HPB) function, and (f) other more subtle changes (e.g. muted responses to oxidative stress, reduced expression of regulatory growth genes, diminished rates of DNA repair, and telomere shortening) [14]. These changes may contribute to the reduced hepatic regenerative capacity in the elderly [14], and this may compound the risk associated with major hepatectomy and advanced age. It has been reported that liver resections for HCC and colorectal liver metastases are as safe in the elderly as those in younger patients [5,15,16]. However, few studies have focused on the outcomes after major hepatectomy (resection of three or more Couinaud segments) in patients with liver cirrhosis aged over 70 years. DOI: 10.1097/MEG.0000000000000046
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Major hepatectomy for HCC in elderly patients Wang et al.
In this study, the outcome of patients with HCC and concomitant liver cirrhosis who underwent major hepatectomy was studied retrospectively, and the influence of age on early postoperative recovery was evaluated. The aim was to compare the rates and severity of postoperative complications following major liver resection in two different age groups.
Patients and methods This retrospective pair-matched study was carried out after obtaining institutional review board approval from the Medical College of Xi’an Jiaotong University. Written consent was given by patients for their information to be stored in the hospital database and used for research. All treatment strategies and indications for surgery were discussed by a multidisciplinary board. Before surgery, all patients had been informed of the necessity of surgery and the potential for perioperative complications, they all agreed to surgery and gave informed consent. Patients aged 70 years or more (group E) who underwent major hepatectomy between 2007 and 2012 were matched with those aged less than 70 years (group Y) who underwent major hepatectomy, according to diagnosis and the extent of liver resection. The study included only HCC patients with concomitant liver cirrhosis who underwent major hepatectomy. Patients younger than 18 years, those with noncirrhotic HCC, and those who had a minor resection were excluded from the analysis. Postoperative morbidity and mortality was defined by the Clavien–Dindo classification system [17] and liver failure (LF) by the ‘50–50’ criteria at postoperative day 5 [18]. Postoperative mortality included death within 90 days of liver resection. The Clavien–Dindo classification of surgical complications within 30 days of the postoperative period was also used to grade the severity of postoperative complications. Postoperative hemorrhage was considered when bleeding from the drainage tubes exceeded 100 ml/h. The diagnostic criteria of pneumonia are based on imaging and sputum culture, and bacteremia are based on blood culture. Patient demographics, diagnoses, preoperative condition, surgical procedures, and postoperative outcomes were systematically recorded in different groups, and the data were analyzed retrospectively. Preoperative, intraoperative, and postoperative data were collected prospectively in a computerized database. The data were imported into SPSS software (version 16.0; SPSS Inc., Chicago, Illinois, USA) for analysis. Univariate w2 and Fisher’s exact test (in cases of low frequency) were performed to discriminate discrete variables. The data were non-normally distributed, and therefore nonparametric statistical analyses were used. Mann–Whitney U-tests were performed to test the differences between groups with continuous variables, and w2 analysis was
445
performed for categorical variables. Significance was set at P value less than 0.05. Univariate analysis and multivariate analysis were performed to determine possible predictors of complications in elderly patients.
Perioperative management
Before surgery, a senior HPB surgeon and an anesthetist assessed all patients. Preoperative investigation of the patients included blood biochemistry, a-fetoprotein assay, chest radiograph, ECG, percutaneous ultrasonography, high-resolution contrast-enhanced computed tomography scan of the chest and abdomen, gastroscope, and/or indocyanine green clearance test. The aim of imaging was to assess the liver volume and future remnant liver volume, vascular anatomy, portal hypertension, the extent of the hepatic lesion and to rule out extrahepatic malignant disease. The diagnosis of cirrhosis was made according to the guidelines published by Desmet et al. [19]. All cases were pathologically confirmed HCC with concomitant cirrhosis. All patients’ future remnant liver volume was more than 40%. Liver function was assessed by Child–Pugh classification and/or indocyanine green retention test. All patients older than 65 years were assessed for cardiac risk by a dedicated preoperative workup according to the guidelines of European Society of Anesthesiology (ESA) [20]. Liver resections were classified according to the Brisbane nomenclature [21]. Intermittent vascular occlusion was performed with cycles of 15-min inflow occlusion followed by 5 min of reperfusion. Central venous pressure was kept below 5 mmHg during liver parenchymal transection. Large vessels and bile ducts were ligatured with a suture, clip or vascular stapling device or secured intrahepatically with interrupted sutures. Close-suction drainage tube was selectively placed in the subphrenic or subhepatic space close to the transection surface at the end of the operation according to intraoperative conditions, and was unplugged at the most opportune time. All patients were nursed in the ICU during the early postoperative period to maintain the fluid balance, and for cardiac monitoring and respiratory management. Postoperative liver and kidney function and routine blood tests were assessed. On the basis of the liver function tests, some patients were treated with albumin or fresh frozen plasma to maintain postoperative albumin levels above 35 g/l to avoid ascites or pleural effusion. Early oral feeding was encouraged and parenteral nutrition was started immediately after major hepatectomy for all patients. Careful attention was paid to the management of intravenous fluid and electrolyte balance during the early postoperative period. Intravenous fluid was rigidly restricted to less than 2 l/day to minimize sodium and fluid retention and reduce the potential for ascites accumulation in all patients. Transcutaneous ultrasonography was performed on patients on postoperative day 3 or 7 by
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
446 European Journal of Gastroenterology & Hepatology 2014, Vol 26 No 4
Table 1 Preoperative characteristics of patients undergoing major hepatectomy for hepatocellular carcinoma with concomitant liver cirrhosis Variables Median age (range) (years) Sex (male/female) HBV-related cirrhosis [n (%)] HCV-related cirrhosis [n (%)] Diameter of tumor (cm) Number of tumor History of abdominal surgery [n (%)] History of alimentary tract hemorrhage [n (%)] History of TACE [n (%)] Smoking [n (%)] Comorbidities Hypertension Diabetes Cardiac arrhythmias COPD Other Platelet count ( 109/l) Hemoglobin (g/l) White blood cell count ( 109/l) Prothrombin time (s) ALT (IU/l) AST (IU/l) Total bilirubin (mmol/l) Total cholesterol (mmol/l) Albumin (g/l) Serum creatinine Serum sodium Serum AFP (ng/ml) Child–Pugh grade (n) Class A Class B ASA grade (n) Grade 1 Grade 2 Grade 3
Group E (Z 70 years; n = 56) 74 (70–78) 40/16 54 (96.42) 2 (3.58) 7.50 (3.00–15.00) 1.18±0.61 14 (25.00) 8 (14.29) 8 (14.29) 16 (28.57)
Group Y (< 70 years; n = 152) 48 (19–69) 120/32 150 (98.68) 2 (1.32) 8.85 (2.40–17.50) 1.20±0.56 24 (15.79) 12 (7.89) 16 (10.53) 52 (34.21)
20 4 6 4 0 119.00 (36.00–500.00) 124.46±18.48 4.71 (1.80–70.00) 13.02±1.46 35.50 (11.00–184.00) 40.50 (17.00–205.00) 16.95 (4.62–338.90) 3.86±1.04 38.62 (29.09–47.65) 74.66 (62.58–105.79) 140.72 (131.62–144.70) 97.02 (2.47–31104.00)
16 4 6 8 3 135.50 (17.00–348.00) 127.32±16.55 5.12 (1.40–20.29) 13.36±1.37 45.50 (13.00–414.00) 52.00 (12.00–619.00) 15.77 (5.05–378.92) 3.76±1.22 38.72 (15.62–47.60) 75.26 (44.69–117.70) 139.52 (132.00–146.50) 1210 (1.89–121000.00)
48 8
122 30
0 38 18
10 92 50
P value < 0.001 0.254 0.630 0.630 0.122 0.014 0.127 0.165 0.452 0.442 < 0.001 0.274 0.128 0.857 0.687 0.946 < 0.001 0.759 < 0.001 0.043 0.035 0.714 < 0.001 0.626 0.159 0.611 0.107 0.367
0.113
AFP, a-fetoprotein; ALT, alanine aminotransferase; AST, aspartate aminotransferase; ASA, American Society of Anesthesiologists; COPD, chronic obstructive pulmonary disease; HBV, hepatitis B virus; HCV, hepatitis C virus; TACE, transcatheter arterial chemoembolization.
experienced radiologists to detect any perihepatic or celiac collection of ascitic fluid or pleural effusion.
Results Demographics and perioperative details
There were 1018 consecutive liver resections performed between 2007 and 2012 at the Department of Hepatobiliary Surgery, First Affiliated Hospital of Medical College, Xi’an Jiaotong University, Xi’an, China. Among these patients, a total of 208 histologically confirmed HCC patients with concomitant cirrhosis underwent major hepatectomy. We identified 56 patients aged 70 years or more with major liver resection who could be matched with 152 control patients aged below 70 years undergoing similar types of liver resection. Demographics, preoperative comorbidities, and preoperative laboratory values are shown in Table 1. The male-tofemale ratio was 40 : 16 and 120 : 32 in groups E and Y, respectively (P = 0.254). The median age was 74 years (range 70–78 years) and 48 years (range 19–69 years) in groups E and Y, respectively. Most patients had hepatitis B virus (HBV)-related cirrhosis (96.42% in group E vs. 98.68% in group Y; P = 0.630). A higher rate of hypertension was reported in group E (P < 0.001). There were also
significant differences in terms of the number of tumors (P = 0.014), preoperative hemoglobin (P < 0.001), prothrombin time (P < 0.001), and preoperative total cholesterol (P < 0.001), favoring group Y. However, there were significant differences in terms of alanine aminotransferase (ALT) (P = 0.043) and aspartate aminotransferase (AST) (P = 0.035), favoring group E. There were no differences in terms of other clinical variables. Intraoperative and postoperative results for both groups are shown in Table 2; there were no significant differences between the two groups. Postoperative changes in serum ALT, AST, total bilirubin (TBIL) and albumin (ALB) in the two groups are shown in Fig. 1. The changes in ALT, AST, and ALB were not significantly different between the two groups (P = 0.903, P = 0.791, and P = 0.594, respectively). However, there were significant differences in TBIL between the two groups (P = 0.026). There was no difference in postoperative albumin infusion between the two groups (P = 0.353). Postoperative complications
Postoperative complications in both groups are shown in Table 3. The mortality rate was 3.57% (two deaths) in
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Major hepatectomy for HCC in elderly patients Wang et al. 447
Table 2
Risk factor analysis
Intraoperative or postoperative results
Variables
Group E (Z 70 years; n = 56)
Group Y (< 70 years; n = 152)
Type of resection [n (%)] Right hepatectomy 24 (42.86) 44 (28.95) Left hepatectomy 10 (17.86) 38 (25.00) Extended right 0 (0.00) 2 (1.32) hepatectomy Extended left hepatectomy 0 (0.00) 6 (3.95) Resection of three 22 (39.29) 62 (40.79) segments Simultaneous extrahepatic procedures [n (%)] Diaphragm resection/repair 6 (10.71) 4 (2.63) Bile duct excision 4 (7.14) 12 (7.89) Cholangiojejunostomy 6 (10.71) 4 (2.63) Other upper 2 (3.57) 2 (1.32) gastrointestinal resection Portal vein tumor thrombus 14 (25.00) 22 (14.47) Splenic artery ligation 2 (3.57) 4 (2.63) Portal vein reconstruct 16 (28.57) 26 (17.11) Other 2 (3.57) 6 (3.95) Simultaneous ablation 8 (14.29) 12 (7.89) Intraoperative blood loss (ml) 950 (200–4000) 850 (100–7500) Intraoperative RBC 4 (2–18) 4 (0–24) transfusion (U) Hepatic portal occlusion (min) 14.0 (0.00–54.00) 15.50 (0.00–58.00) Operative time (min) 245 (140–490) 280 (90–480) Postoperative albumin 90 (0–450) 75 (0–380) transfusion (g) ICU stay (days) 3 (2–6) 3 (1–11) Postoperative hospital stay 15 (6–32) 15 (5–42) (days)
P value 0.225
0.563
On univariate analysis, intraoperative blood loss of 1000 ml or more (P = 0.021), intraoperative red blood cell (RBC) transfusion of 5 U or more (P = 0.038), and operative time of 250 min or more (P = 0.027) were found to be predictive of postoperative complications (Table 4). In multivariate analysis, intraoperative RBC transfusion of 5 U or more (P = 0.016; hazard ratio 4.812; 95% CI 1.332–17.384) was a predictor of higher morbidity in the elderly (Table 5). In addition, patients with complications had a longer stay in the ICU (P = 0.015) and in hospital (P = 0.043) than those without complications.
Discussion
0.165 0.713 0.728 0.376 0.113 0.353 0.480 0.975
RBC, red blood cell.
group E compared with 1.32% (two deaths) in group Y, and this was not significantly different [P = 0.630; odds ratio (OR) 2.778; 95% confidence interval (CI) 0.382–20.209]. In both groups, the causes of death were LF. Overall complication rates were 53.57% in group E and 47.37% in group Y; this difference was not significant (P = 0.427; OR 1.282; 95% CI 0.694–2.369). A total of 25.00% of patients experienced complications requiring invasive interventions (Clavien–Dindo classification grade Z 3a) in group E compared with 13.16% in group Y (P = 0.040; OR 2.200; 95% CI 1.023–4.733). A number of postoperative complications were more common in group E including: postoperative pneumonia (P < 0.001), bacteremia (P = 0.026), and respiratory failure requiring reintubation (P = 0.028). Two elderly patients had postoperative confusion related to liver insufficiency, and they recovered with conservative treatment. With regard to liver-related complications, there was one bile leak in an elderly patient, which required percutaneous drainage. There were four cases of bile leak in group Y, two received endoscopic nasobiliary drainage, one required percutaneous drainage, and one required percutaneous drainage of biloma. None of the patients died due to bile leakage. In addition, LF occurred in four patients in group E and six in group Y, and there was no significant difference (P = 0.555).
An aging worldwide population has resulted in a significant increase in the number of elderly patients considered for resection of malignant HPB tumors. Malignant HPB tumors most commonly occur during the sixth to eighth decades of life [22]. In addition, the majority of patients with HCC have underlying cirrhosis [11,12]. Previous studies have showed that liver resection is associated with high morbidity and mortality rates, especially in patients with cirrhosis [23–25]. With the current improvements in surgical techniques and perioperative care, major hepatic resection for HCC, mainly in the form of classical right or left hepatectomy, is associated with acceptable operative morbidity and mortality in selected patients with cirrhosis [11,26,27]. Some studies also demonstrate that major hepatectomy can be performed as safely in elderly patients as in younger patients [5,15,16,28]. However, there is a paucity of data in the literature on major resection for HCC in cirrhotic elderly patients. There are a number of biases against major hepatectomy for elderly HCC patients with concomitant liver cirrhosis including: (i) elderly patients have less aggressive tumors compared with their younger counterparts [29]; (ii) an increased incidence of medical comorbidities leads to a reduced tolerance to surgery [30]; and (iii) aging leads to a number of structural and functional changes in the liver that may make liver resection less tolerable [14]. Furthermore, despite recent advances in surgical techniques and perioperative management, hepatic resection in the cirrhotic liver remains a major challenge for surgeons because of the specific risks associated with cirrhosis, such as difficult mobilization and transection of the fibrotic liver, increased bleeding as a result of portal hypertension, and higher risk of LF [30]. Resection of HCC in cirrhotic liver may be fruitless as there is a high risk of recurrence in the remnant cirrhotic liver [4,23,24]. Local therapy for small HCC, such as percutaneous alcohol injection, transcatheter hepatic arterial chemoembolization, and radiofrequency ablation, gives long-term results similar to those of surgery [31]. Some studies have revealed that elderly HCC patients with cirrhosis after major liver resection derive as much
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
448 European Journal of Gastroenterology & Hepatology 2014, Vol 26 No 4
Fig. 1
(a)
(b) 600.00
P = 0.903
P = 0.791 600.00 AST in U/l (median)
ALT in U/l (median)
500.00 400.00 300.00 200.00
400.00
200.00
100.00 0.00
0.00 PO
POD1
POD3
POD7
(c) P = 0.026
60.00
POD1
POD3
POD7
50.00 40.00 30.00
P = 0.594
45.00
ALB in g/l (median)
TBIL in μmol/l (median)
PO (d)
20.00 10.00
40.00
35.00
30.00
25.00 PO
POD1
POD3
POD7 Group E
PO
POD1
POD3
POD7
Group Y
Serum levels of alanine aminotransferase (ALT), aspartate aminotransferase (AST), total bilirubin (TBIL), and albumin (ALB). (a) Median perioperative changes in levels of ALT in U/l, (b) AST in U/l, (c) TBIL in mmol/l, and (d) ALB in g/l. The smallest and the largest observations are represented by the ends of the error bars (5th percentile and 95th percentile, respectively). The horizontal solid lines represent the normal values for each parameter. There was no significant difference in preoperative and postoperative serum levels of ALT, AST, and ALB between the groups. There was a significant difference in preoperative and postoperative serum levels of TBIL between the groups (P = 0.026). PO, preoperation; POD, postoperative day.
benefit as their younger counterparts [12,27,32]. In the current study, the 53.57% morbidity and 3.57% mortality rates in elderly HCC patients with cirrhosis are comparable with the rates reported in the published literature [4,11,27]. Our study demonstrates that major liver resection can be performed in HCC patients with cirrhosis aged 70 years or more with comparable early outcomes to those observed in younger patients. Advanced age is not the major determinant of the morbidity and mortality rates associated with major liver resection in elderly HCC patients with cirrhosis. This provides important evidence to dispel the common misconception that all elderly HCC patients with cirrhosis belong to a high surgical risk group by dint of their advanced age alone. Therefore, each elderly HCC patient with cirrhosis should be assessed for individual risk, and advanced age
should not be considered a contraindication for major liver resection. It is well established that the function of most organs usually deteriorates with age. In this study, the incidence of hypertension, and preoperative total cholesterol levels were significantly higher, and hemoglobin level was significantly lower in group E compared with group Y. This may reflect a decreased liver reserve capacity in the elderly. However, these factors did not result in an increase in mortality and morbidity rates in the elderly. Preoperative ALT and AST were significantly higher in group Y than in group E. This may reflect the aggressive selection of patients in group Y and prudent selection of patients in group E. Furthermore, because there was no significant difference between the two groups in the American Society of Anesthesiologists
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Major hepatectomy for HCC in elderly patients Wang et al. 449
Table 3
Details of postoperative complications
Complications Postoperative morbidity [n (%)] Postoperative mortality [n (%)] Confusion Wound infection Pneumonia Pleural effusion Pleural effusion requiring drainage Ascites Ascites requiring drainage Bile leak Hemorrhage Bacteremia Arrhythmia Liver failure Other organ failure Reoperation Respiratory failure requiring reintubation MODS Clavien–Dindo classification grade of complications No complication Grade 1 Grade 2 Grade 3a Grade 3b Grade 4a Grade 4b Grade 5
Group E (Z 70 years; n = 56)
Group Y (< 70 years; n = 152)
P value
30 (53.57) 2 (3.57) 2 4 12 8 4 24 18 1 4 3 6 4 2 0 4
72 (47.37) 2 (1.32) 0 4 6 16 6 48 36 4 12 0 5 6 1 1 1
0.427 0.630 0.123 0.274 < 0.001 0.452 0.555 0.129 0.217 0.999 0.999 0.026 0.076 0.555 0.364 0.999 0.028
4
2
26 4 22 6 0 6 2 2
80 14 64 12 2 6 2 2
0.078 0.389
MODS, multiple organ dysfunction syndrome.
(ASA) grade, the severity of associated diseases may be comparable between the groups. Although the recovery of postoperative serum TBIL level was inferior in elderly patients, they did not have higher liver insufficiencyrelated morbidity than younger patients. Elderly patients had similar changes in postoperative ALT, AST, and ALB values when compared with the younger patients. In view of these findings, we could not arbitrarily find any evidence that elderly patients were more susceptible to liver insufficiency than younger patients. Our study is noteworthy in that we have recorded all complications in the postoperative period using the Clavien– Dindo complication classification grade system, whereas many of the published studies have only reported complication rates [4,11,27]. For those studies that confuse the types of complications, treatment interventions and therefore the severity of complications were not uniformly disclosed, suggesting our complication rate may be higher than other reports. Of note, older age was associated with more severe complications (Z grade 3a) requiring invasive procedures. Therefore, this suggests that while elderly patients require more intense monitoring and greater nursing care and physical therapy, they can expect similar surgical outcomes. In a multivariate analysis, intraoperative RBC transfusion of 5 U or more (P = 0.016; hazard ratio 4.812; 95% CI 1.332–17.384) was a predictor of higher morbidity in the elderly. Bernard et al. [33] concluded that intraoperative
transfusion of 1–2 U packed RBCs is associated with increased 30-day mortality, surgical-site infection, pneumonia, and sepsis in general surgery patients. Therefore, intraoperative blood transfusion must strictly abide by transfusion guidelines. Interestingly, however, time of hepatic portal occlusion, simultaneous extrahepatic procedures and number of Couinaud liver segments resected did not correlate with complication rates, suggesting that the complexity of surgery itself was not responsible for the higher complication rates in the elderly. Overall, 75–80% of HCC is attributable to chronic HBV (50–55%) or hepatitis C virus (25–30%) infection [34–37]. Furthermore, HBV infection may be higher in Asia [11]. In our study, all patients had HCC with cirrhosis, indicating that HBV infection was more common. These patients had better liver function and fewer tumors than younger patients. Therefore, these patients may be good candidates for major liver resection. It should be noted, however, that numerous studies have reported that resection for HCC can reactive HBV, initiate virus replication, and even result in LF or cancer recurrence. Therefore, antiviral therapy seems to be an appropriate option, and should even be a part of the protocol for all HBV-related HCC patients who undergo liver resection, irrespective of whether the viral loads are high or low [38,39]. Unfortunately, in the current study there is a lack of data on antiviral treatment and patients were not followed for longer observation periods. Therefore, we could not determine the benefit of antiviral treatment in elderly HBV-related HCC patients with cirrhosis. This study shows that major hepatectomy can be performed in HCC patients aged 70 years or more with concomitant liver cirrhosis with comparable outcomes to those observed in younger patients. However, our study is limited by the small sample size, a retrospective analysis, rigorous selection criteria, and a lack of antiviral treatment data. The statistical power may be insufficient to detect a small susceptible effect of some factors on univariate and multivariate analysis. Therefore, prospective studies in larger samples of patients with aggressive surgical procedures (such as extended right/left hepatectomy) and longer observation periods are needed to clarify the effects of age on elderly HCC patients with cirrhosis. In conclusion, major liver resection can be relatively safely performed in carefully selected elderly HCC patients with concomitant liver cirrhosis. Advanced age alone should not be considered a contraindication for major liver resection. With improving perioperative assessment and operative techniques, most complications after cirrhotic liver resection can be treated and have a low mortality rate. However, more care should be taken if intraoperative RBC transfusion of more than 5 U is required. Surgeons should take care to minimize the likelihood of intraoperative blood transfusion in elderly patients.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
450
European Journal of Gastroenterology & Hepatology 2014, Vol 26 No 4
Table 4
Univariate analysis of variables predictive of complications in elderly patients
Covariables
Affected by complications (%)
Sex Male Female Age History of abdominal surgery History of alimentary tract hemorrhage History of TACE Smoking Hypertension Diabetes Cardiac arrhythmias COPD Platelet count (109/l) White blood cell count ( 109/l) Hemoglobin (g/l) Prothrombin time (s) Preoperative ALT (IU/l) Preoperative AST (IU/l) Preoperative total bilirubin (mmol/l) Total cholesterol (mmol/l) Serum creatinine Preoperative albumin (g/l) Serum sodium Serum AFP (ng/ml) > 400 (%) Child–Pugh grade Class A Class B ASA grade (%) Grade 1 Grade 2 Grade 3 Type of resection (%) Right hepatectomy Left hepatectomy Number of liver segment resected Z4 3 Simultaneous extrahepatic procedures Simultaneous ablation Intraoperative blood loss (ml) Z 1000 < 1000 Intraoperative RBC transfusion (U) Z5
