This was a retrospective case–control study. We consecutively reviewed 1,452 eligible patients who underwent hepatectomy between January 2013 and December 2016 in the Hepatic Surgery Center of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology. According to the imaging, including gallbladder ultrasound, computerized tomography (CT), and magnetic resonance imaging (MRI), all patients were divided into the gallstone group and the nongallstone group. The flow chart of patients is displayed in Figure 1.

The inclusion criteria are as follows: (1) Child–Pugh A or B; (2) primary liver cancer Barcelona Clinic Liver Cancer (BCLC) A/B stages; and (3) benign tumor of liver including hepatic adenoma, hepatic hemangioma, and hepatic focal nodular hyperplasia (FNH). The exclusion criteria are as follows: (1) metastatic liver cancer; (2) history of cholecystectomy; (3) already existing cholecystitis or gallstones; (4) intrahepatic bile duct stones; (5) consistent use of hormone replacement therapy; and (6) abdominal radiotherapy.

The gallbladder is attached to S5 of the liver. In general, anatomical hepatectomy usually combines with cholecystectomy when a tumor is located in S5 of the liver. When the tumor is small and nonanatomic hepatectomy is performed, gallbladder preservation is feasible. Thus, there is a considerable number of patients with tumors located in S5 of the liver in our study who underwent hepatectomy with gallbladder preservation.

Postoperative complications are common after hepatectomy. The severity of complications was classified according to the Clavien–Dindo classification system (9). Grade I complications included mild pleural, peritoneal effusion, electrolyte disturbance, and other deviations from the normal postoperative course. Grade II complications included blood transfusion, plasma infusion, or albumin infusion. Grade III covered severe pleural and peritoneal effusion requiring percutaneous aspiration or drainage. Grade IV complications covered severe lung infections, respiratory failure, or kidney failure, requiring intubation or dialysis treatment.

Hepatectomy was accomplished in a hepatic surgery center, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology. All patients were given general anesthesia. The methods of hepatectomy included open and laparoscopic hepatectomy. Partial resection or anatomic resection of the liver was performed based on the tumor location. Complete removal of at least one Couinaud segment containing the focus and the portal vein in the drainage area of the lesion was defined as an anatomic resection. A complete tumor plus a rim of non-neoplastic liver parenchyma was considered a nonanatomic resection. Partial resection makes preservation of the gallbladder possible when the surgeon removes the small lesions located at S5.

The study was approved by the Ethical Committee of Tongji Hospital of Tongji Medical College of Huazhong University of Science and Technology. All procedures performed in this study abided by the Declaration of Helsinki.

Continuous variables were presented as the median and interquartile range (IQR), and categorical data as were presented numbers and percentages. A χ² test and the Mann–Whitney U test were used to compare groups where appropriate. Univariate and multivariate logistic regression analyses were performed to evaluate risk factors for gallstone development. P < 0.05 was considered statistically significant. Statistical analyses were performed using SPSS version 19.0 for Windows (SPSS, Chicago, IL, United States).

Figure 2 shows that 1,452 eligible patients were stratified according to the malignant or benign tumor. In the total sample of included patients, there were 981 patients with primary liver cancer and 471 patients with benign tumors, including 24 patients with hepatic adenoma, 387 patients with hepatic hemangioma, and 60 patients with FNH.

The median age was 56 years, and 85.5% (1,241/1,452) of patients were men. In total, 96.6% (1,403/1,452) of patients were Child–Pugh A. The etiology of patients was 776 (53.4%) positive for hepatitis B virus (HBV), 18 (1.2%) positive for hepatitis C virus (HCV), 41 (2.8%) for other causes, and 617 (42.6%) with no hepatitis background. A higher body mass index (BMI), total bilirubin, tumor size, and lower albumin were observed in the gallstone group. The proportion of the female gender, diabetes, Child–Pugh B, hepatitis B surface antigen (HBsAg)-positive, liver cirrhosis, S5 liver resection, multiple tumor, and postoperative complications was higher in the gallstone group (Table 1).

The overall gallstone incidence was 23.5% (341/1,452), and the asymptomatic gallstone incidence was 83.6% (285/341). The incidence of gallstones in the primary liver cancer group was 25.7% (252/981) higher than that in the benign tumor group [18.9% (89/471), P = 0.004]. Asymptomatic gallstones accounted for 81.7% (206/252) of all gallstones in the primary liver cancer group and 89.2% (79/89) of all gallstones in the benign tumor group (P = 0.11).

In 4 years, 41 out of 341 patients (12.0%) with gallstones underwent cholecystectomy. All patients with cholecystectomy had significant right upper abdominal pain, and their Murphy sign was positive. In addition, CT indicated thickening of the gallbladder wall and chronic cholecystitis. No cholecystectomy was performed for all asymptomatic and some symptomatic gallstones. Moreover, we had not observed any serious disease caused by gallstones, including acute pancreatitis and gallbladder cancer.

Univariate and multivariate logistic regression analyses showed that female gender (OR: 1.547, 95% CI: 1.358–2.356, P = 0.018), BMI (OR: 3.255, 95% CI: 2.789–4.589, P = 0.009), liver resection at S5 (OR: 3.687, 95% CI: 2.987–4.698, P = 0.012), and postoperative complication (OR: 3.684, 95% CI: 3.489–4.398, P < 0.001) were factors related to gallstone in patients with benign liver tumors after hepatectomy (Table 2).

Univariate and multivariate logistic regression analyses showed that female gender (OR: 2.314, 95% CI: 2.021–4.529, P = 0.037), BMI (OR: 2.125, 95% CI: 1.546–3.356, P = 0.011), diabetes (OR: 1.846, 95% CI: 1.259–2.136, P = 0.019), Child–Pugh B (OR: 2.472, 95% CI: 1.261–4.025, P = 0.008), albumin (OR: 2.851, 95% CI: 2.622–3.914, P < 0.001), liver cirrhosis (OR: 4.258, 95% CI: 3.258–6.398, P < 0.001), liver resection at S5 (OR: 4.589, 95% CI: 3.514–6.215, P < 0.001), postoperative complication (OR: 6.256, 95% CI: 5.021–7.654, P < 0.001), and TACE after recurrence (OR: 2.691, 95% CI: 2.102–3.028, P = 0.019) were factors related to gallstones in patients with primary liver cancer after hepatectomy (Table 3).

Figure 3 shows three typical cases of gallstone formation after hepatectomy or recurrent patients receiving TACE. Case 1 displayed that the tumor was located in S5 (white triangle) and there was no gallstone in the gallbladder before surgery (Figures 3A,B). The gallstone developed in the gallbladder (arrow), and the gallbladder tightly adhered to the area of abdominal adhesions (Figure 3C). Case 2 showed that the surgical area was located in S4 and partial S5, and the surgical area presented with effusion and adhesions (arrow) (Figures 3D,E). The gallstone developed in the gallbladder ultimately (arrow) (Figure 3F). Case 3 showed that a recurrent patient after surgery whose lesion was located in S5 received TACE treatment; most areas of the lesion are necrotic (white triangle). The gallstone developed in the gallbladder ultimately and the wall of the gallbladder was significantly thickened (arrow) (Figures 3G,H).