Liver resection for hepatocellular carcinoma in oldest old patients

We undertook a retrospective review and analysis of the clinicopathological data of 1910 consecutive patients who had undergone liver resections for newly diagnosed HCC during the period from January 1992 to December 2016. Patients whose HCC resection needed cardiopulmonary bypass for tumor thrombus extending to the right atrium (n = 11) [16] and patients who received non-curative liver resection (n = 10, defined as gross residual tumor after operation [8]) were excluded.

Among all enrolled patients, 1858 of them were aged < 85 years at the time of operation (group A, median 62, range 18–84). The age of the remaining 31 patients was ≥ 85 years (group B, median 87.5 years, range 85–95). We then compared groups A and B regarding patients’ background features, tumor characteristics, and the early and long-term postoperative outcomes.

During the same study period, 18 oldest old patients were judged as resectable for HCC but were treated by non-operative modalities instead due to the refusal of operation by the patients or their families (group C). Their treatment modalities were RFA in 4, TACE in 9, oral sorafenib in 3, and conservative treatment in 2. TACE or RFA was performed by a senior radiologist (JIH).

Since July 2012, laparoscopic hepatectomy was performed on one patient in group B and 51 in group A. Tumors of these patients met the patient selection criteria of the Louisville statements of laparoscopic hepatectomy [17].

As we have previously reported [8], HCC patients underwent measurement of conventional hemogram and liver function tests: serum α-fetoprotein (AFP), hepatitis B surface antigen (HBsAg), and anti-hepatitis C antibody (anti-HCV); indo-cyanine green (ICG) clearance test; gastroduodenal endoscopy; abdominal computed tomography (CT); and/or magnetic resonance imaging (MRI). The resectability and the extent of the liver to be resected were based on the tumor extension and modified Makuuchi criteria [18].

Patients whose co-morbidities should be well treated and controlled and who fell into American Society of Anesthesiology (ASA) class one or two were considered operable. After 2001, forced expiratory volume at 1 s (FEV 1) ≥ 75% and left ventricular ejection fraction (LVEF) in echocardiogram ≥ 50% were added to the patient selection criteria in patients older than 65 years. The treatment strategies, management plans, and tentative operative procedures of each patient were agreed upon before operation in conferences attended jointly by surgeons, gastroenterologists, anesthesiologists, radiologists, and co-morbidity-related physicians. The associated co-morbidities were the following: cardiopulmonary diseases, including hypertension, heart failure, cardiomyopathies, valvular heart disease, pericardial disease, syncope, aortic aneurysms, coronary arterial disease, and arrhythmia; lung malignancies, including chronic obstructive pulmonary disease, bronchiectasis, obstructive sleep apnea, interstitial lung disease, and pulmonary hypertension; neuromuscular disorders, including seizure, myopathies, stroke, dementia, and Parkinson’s disease; gastrointestinal diseases, including peptic ulcer disease, malignancies other than HCC, inflammatory bowel disease, gastrointestinal bleeding, biliary tract diseases, esophageal and gastric disorders; endocrine and metabolic diseases, including breast disease, thyroid disease, adrenal disorders, pituitary disorders, diabetes mellitus, and dyslipidemia; urologic and nephrologic diseases, including end-stage renal disease, urologic and genital malignancies, lithiasis, renal failure and glomerular disease; hematologic diseases, including anemia, platelet disorders, coagulopathies, leukemias, lymphomas and myeloproliferative disease; and rheumatologic diseases, including systemic lupus erythematosus, vasculitis, amyloidosis, various arthritis, and other autoimmune diseases. For patients with end-stage renal disease, peri-operative heparin-free hemodialysis was carried out [19]. For patients with hypersplenic thrombocytopenia (platelet count ≤ 80,000/mm³), concomitant splenectomy was carried out [20]. For patients with severe gastroesophageal varices (F3 or presence of red-colored sign), we performed pre-operative endoscopic ligation and/or sclerotherapy [21]. In addition to post-operative pain control, epidural catheter insertion for analgesic injection was performed routinely after 2005.

Liver parenchyma was transected using a Kelly crushing method under intermittent hepatic inflow blood occlusion (Pringle’s maneuver). During liver parenchymal transection, a low central venous pressure (CVP) policy (CVP < 5 cm H₂O) was implemented by a senior anesthesiologist (CHS), who also oversaw the restrictive blood transfusion policy.

Resected specimens were examined by a senior pathologist (YGJ), who determined the following: tumor capsular formation, resection margin width, tumor number, micro- or macro-vascular invasion, Ishak score, cirrhosis severity [22], and tumor differentiation (using Edmondson and Steiner grading). The AJCC Cancer staging system (8th edition) was applied after pathological examinations. Early postoperative complications, such as bile leakage, ascites, and liver failure, were recorded as defined according to the international consensus [23‐25]. The severity of complications [26] was determined using the Clavien-Dindo classification. Any death that had occurred within 90 days after the operation was considered operative mortality.

Patients who survived from hepatectomy were followed up at the outpatient clinic during the first 2 years at intervals of 2 to 3 months, and thereafter at intervals of 4–6 months. Serum AFP, liver function, and imaging (abdominal ultrasonography, CT, or MRI) were checked. Recurrent HCCs were treated by re-hepatectomy, RFA, TACE, oral sorafenib, or other conservative treatments, as appropriate.

Continuous variables are expressed as median (range) and were compared using Mann-Whitney U test. Frequencies were compared using Fisher’s exact test or Pearson’s χ² test. All patients were followed up till July 2018. The disease-free survival rates and overall survival rates were characterized by the Kaplan-Meier life-table method and compared using the log-rank test. p values < 0.05 were considered statistically significant.

Table 1 shows the clinicopathological characteristics of patients with resected HCC (groups A and B) and clinical features of the oldest old HCC patients without surgery (group C). Groups B and C had significantly higher incidences of associated co-morbidities and anti-HCV positivity compared with group A. Clinicopathological characteristics did not significantly differ between groups B and C.

Group B had a significantly longer postoperative hospital stay compared with group A [18.0 (8–46) days vs. 10.0 (7–81) days, p = 0.001]. No significant differences were found in intra- or early post-operative results between the two operated groups (Table 2).  The pathologic features between groups B and A have no significant difference (Table 3).

The difference in DFS between groups B and A was not significant (p = 0.163) (Fig. 1). The OS of Group B was slightly but not significantly lower than that of group A (p = 0.086). On the other hand, group C had very poor outcomes compared to group B (Fig. 2).