Prognostic value of ABO blood groups in upfront operated pancreatic ductal adenocarcinomas.

PURPOSE: Pancreatic ductal adenocarcinoma (PDAC) has been shown to have a lower incidence in patients with blood group O. It is currently uncertain if patients with group O have a better prognosis after pancreatectomy. This study assessed the overall survival (OS) and disease-free survival (DFS) of PDAC patients who underwent upfront pancreatoduodenectomy based on ABO blood groups. METHODS: A cross-sectional study was performed including patients from two university centers. All consecutive head PDAC patients who underwent upfront pancreatoduodenectomy from 2000 to 2016 were included. OS and DFS were compared between blood groups A, B, AB, and O using Kaplan-Meier curves and log-rank tests. RESULTS: A total of 438 patients were included (215 women, median age 67). Pre- and intraoperative details were comparable between all subgroups. Median OS did not differ between the four blood groups (A: 23 months, 95% CI 18-28; B: 32, 95% CI 20-44; AB: 37, 95% CI 18-56 and O: 26, 95% CI 20-32, p = 0.192). Median DFS were also similar (A: 19 months, 95% CI 15-23; B: 26, 95% CI 19-33; AB: 35, 95% CI 15-55 and O: 22, 95% CI 15-29, p = 0.441). There was no OS difference between O and non-O groups (median: 26 months, 95% CI 20-33 vs. 25 months, 95% CI 20-30, p = 0.773). On multivariable analysis blood groups were not prognostic of OS. Only lymph node involvement, tumor differentiation, and adjuvant chemotherapy were independent prognostic factors. CONCLUSION: OS and DFS were similar between all four blood groups after pancreatoduodenectomy. Independent predictors of OS were associated with tumor characteristics and adjuvant treatment.

Guidelines for Perioperative Care for Liver Surgery: Enhanced Recovery After Surgery (ERAS) Society Recommendations 2022.

BACKGROUND: Enhanced Recovery After Surgery (ERAS) has been widely applied in liver surgery since the publication of the first ERAS guidelines in 2016. The aim of the present article was to update the ERAS guidelines in liver surgery using a modified Delphi method based on a systematic review of the literature. METHODS: A systematic literature review was performed using MEDLINE/PubMed, Embase, and the Cochrane Library. A modified Delphi method including 15 international experts was used. Consensus was judged to be reached when >80% of the experts agreed on the recommended items. Recommendations were based on the Grading of Recommendations, Assessment, Development and Evaluations system. RESULTS: A total of 7541 manuscripts were screened, and 240 articles were finally included. Twenty-five recommendation items were elaborated. All of them obtained consensus (>80% agreement) after 3 Delphi rounds. Nine items (36%) had a high level of evidence and 16 (64%) a strong recommendation grade. Compared to the first ERAS guidelines published, 3 novel items were introduced: prehabilitation in high-risk patients, preoperative biliary drainage in cholestatic liver, and preoperative smoking and alcohol cessation at least 4 weeks before hepatectomy. CONCLUSIONS: These guidelines based on the best available evidence allow standardization of the perioperative management of patients undergoing liver surgery. Specific studies on hepatectomy in cirrhotic patients following an ERAS program are still needed.

Functional, biological, and radiological evaluation of the pancreaticojejunal anastomosis 1 year after pancreatoduodenectomy: a prospective study.

PURPOSE: This prospective study aimed to analyze the functional, biological, and radiological aspects of the pancreatic anastomosis 1 year after pancreatoduodenectomy (PD). METHODS: From 2016 to 2019, patients with PD indication were screened. Questionnaires about pancreas insufficiency, fecal elastase tests, and magnetic resonance imaging (MRI) were performed before and 1 year after PD. RESULTS: Twenty patients were prospectively included. The only difference between pre- and postoperative questionnaires was constipation (less frequent 1 year after PD). Median pre- and postoperative fecal elastase levels were 96 µg/g (IQR 15-196, normal value > 200) and 15 µg/g (IQR 15-26, p = 0.042). There were no significant differences in terms of main pancreatic duct (MPD) size (4, IQR 3-5 vs. 4 mm, IQR 3-5, p = 0.892), border regularity, stenosis, visibility, image improvement, and secondary pancreatic duct dilation before and after secretin injection. All patients but one (2 refused and 2 were lost to follow-up, 15/16, 94%) had a patent pancreaticojejunal anastomosis on 1-year MRI. CONCLUSION: Although median 1-year fecal elastase was significantly lower than preoperatively, suggesting that exocrine secretion was altered, the anatomical outcome as assessed by MRI was excellent showing high patency rate (15/16, 94%) at 1 year. This emphasizes the difference between anatomy and function.

[Novelties on the management of alveolar echinococcosis]. / Nouveautés dans la prise en charge de l'échinococcose alvéolaire.

Alveolar echinococcosis is an endemic parasitosis in Switzerland. This pathology mainly infects the liver and develops similarly as a malignant tumor with its ability to spread into the hepatic parenchyma and its capacity of developing distant lesions via hematogenous dissemination. Treatment is based on complete surgical resection coupled with albendazole treatment. Recently, ex vivo liver resections with auto-transplantation have been shown to be feasible in case of end-stage alveolar echinococcosis. Moreover, new biomarkers such as programmed death-ligand 1 (PD-L1), a protein with immunomodulation property, have shown their potential impact on the treatment and follow-up of patients with alveolar echinococcosis.

L'échinococcose alvéolaire est une parasitose endémique en Suisse. Cette pathologie touche principalement le foie et se développe telle une tumeur maligne, par sa propension à envahir le parenchyme hépatique et par sa capacité à développer des lésions à distance par voie hématogène. Le traitement repose sur une exérèse chirurgicale complète couplée à un traitement d'albendazole. Récemment, des techniques de résection hépatique ex vivo avec auto-transplantation ont montré leur faisabilité en cas d'échinococcose alvéolaire avancée. De plus, de nouveaux marqueurs, comme le programmed death-ligand 1 (PD-L1), protéine jouant un rôle dans l'immunomodulation, ont montré leur potentiel impact pour le traitement et le suivi des patients atteints d'échinococcose alvéolaire.

Trends in pancreatic surgery in Switzerland: a survey and nationwide analysis over two decades.

BACKGROUND: Centralisation of highly specialised medicine (HSM) has changed practice and outcome in pancreatic surgery (PS) also in Switzerland. Fewer hospitals are allowed to perform pancreatic surgery according to nationally defined cut-offs. OBJECTIVE: We aimed to examine trends in PS in Switzerland. First, to assess opinions and expected trends among Swiss pancreatic surgeons in regard of PS practice and second, to assess the evolution of PS performance in Switzerland by a nationwide retrospective analysis. METHODS: First, a 26-item survey among all surgeons who performed PS in 2016 in Switzerland was performed. Then, nationwide data from 1998 to 2018 from all hospitals performing PS was analysed including centre volume, perioperative morbidity and mortality, surgical indications and utilisation of minimally invasive pancreatic surgery (MIPS). The national cut-off for regulatory accredited volume centres (AVC) was ≥ 12. Additionally, an international benchmark definition for high volume (≥ 20 surgeries/year) was used. RESULTS: Among 25 surgeons from 15 centres (response rate 51%), the survey revealed agreement that centralisation is important to improve perioperative outcomes. Respondents agreed on a minimum case load per surgeon or centre. Within the nationwide database, 8534 pancreatic resections were identified. Most resections were performed for pancreatic ductal adenocarcinoma (58.9%). There was a significant trend towards centralisation of PS with fewer non-accredited volume centres (nAVC) (36 in 1998 and 17 in 2018, p < 0.001) and more AVC (2 in 1998 and 18 in 2018, p < 0.001). A significantly higher adjusted mortality after pancreatoduodenectomy (PD) was observed in low-volume compared to high-volume hospitals (OR 1.45 [95% CI 1.15-1.84], p = 0.002) and a similar trend compared among AVC and nAVC (OR 1.25 [95% CI 0.98-1.60], p = 0.072), while mortality after distal pancreatectomy (DP) was not influenced by centre volume. CONCLUSIONS: Over the last two decades, centralisation of PS towards higher-volume centres was observed in Switzerland with a decrease of mortality after PD and low mortality after DP. Further centralisation is supported by most pancreatic surgeons. However, the ideal metric and outcome measures for the allocation of highly specialised medicine need further discussion to allow a fair and outcome-focused allocation.

Predictors of complications after liver surgery: a systematic review of the literature.

BACKGROUND: Numerous potential predictors of adverse outcomes have been reported but their performance and utilization in practice seem heterogenous. This study aimed to systematically review the literature on the role and value of predictors of complications after hepatectomy. METHODS: A systematic review following the PRISMA guidelines was performed. Studies on liver transplant were excluded. Only studies assessing overall or major complications were included. RESULTS: A total of 10'965 abstracts were screened. After application of exclusion criteria, 72 articles including 68'480 patients were included. A total of 72 markers with 48 pre-, 9 intra- and 15 postoperative factors were identified as predictors of complications. Preoperative and intraoperative predictive markers retrieved several times with the highest odds ratios (OR) were ASA score (OR range: 1.3-7.5, significant in 8 studies) and intraoperative need for red blood cell transfusion (OR range: 1.2-17.1, significant in 24 studies), respectively. CONCLUSION: Numerous markers have been described to predict the complication risk after hepatectomy. Because of their intrinsic characteristics, most markers such as ASA score and need for red blood cell transfusion are of limited clinical interest. There is a clear need to identify new biomarkers and to develop scores that could easily be implemented in clinical practice.

Prognostic value of positive histological margins in patients with pancreatic head ductal adenocarcinoma and lymph node involvement: an international multicentric study.

BACKGROUND: Resection margin status and lymph node (LN) involvement are known prognostic factors for patients who undergo pancreatoduodenectomy for pancreatic ductal adenocarcinoma (PDAC). This study aimed to compare overall survival (OS) and disease-free survival (DFS) by resection margin status in patients with PDAC and LN involvement. METHODS: A retrospective international multicentric study was performed including four Western centers. Multivariable Cox analysis was performed to identify prognostic factors of OS and DFS. Median OS and DFS were calculated using Kaplan-Meier curves and compared using log-rank tests. RESULTS: A cohort of 814 PDAC patients with pancreatoduodenectomy were analyzed. A total of 651 patients had LN involvement (80%). On multivariable analysis R1 resection was not an independent factor of worse OS and DFS in patients with LN involvement (HR 1.1, p = 0.565; HR 1.2, p = 0.174). Only tumor size, grade, and adjuvant chemotherapy were associated with OS and DFS. Median OS and DFS were similar between patients with R0 and R1 resections (23 vs. 20 months, p = 0.196; 15 vs. 14 months, p = 0.080). CONCLUSION: Resection status was not identified as predictor of OS or DFS in PDAC patients with LN involvement. Extensive surgery to achieve R0 resection in such patients might not influence the disease course.

Cost Analysis of Enhanced Recovery Programs in Colorectal, Pancreatic, and Hepatic Surgery: A Systematic Review.

BACKGROUND: Enhanced recovery programs (ERPs) have been shown to improve postoperative outcomes after abdominal surgery. This study aimed to review the current literature to assess if ERPs in colorectal, pancreas, and liver surgery induce cost savings. METHODS: A systematic review was performed including prospective and retrospective studies comparing conventional management versus ERP in terms of costs. All kinds of ERP were considered (fast-track, ERAS®, or home-made protocols). Studies with no mention of a clear protocol and no reporting of protocol compliance were excluded. RESULTS: Thirty-seven articles out of 144 identified records were scrutinized as full articles. Final analysis included 16 studies. In colorectal surgery, two studies were prospective (1 randomized controlled trial, RCT) and six retrospective, totaling 1277 non-ERP patients and 2078 ERP patients. Three of the eight studies showed no difference in cost savings between the two groups. The meta-analysis found a mean cost reduction of USD3010 (95% CI: 5370-650, p = 0.01) in favor of ERP. Among the five included studies in pancreas surgery (all retrospective, 552 non-ERP vs. 348 ERP patients), the mean cost reduction in favor of the ERP group was USD7020 (95% CI: 11,600-2430, p = 0.003). In liver surgery, only three studies (two retrospective and 1 RCT, 180 non-ERP vs. 197 ERP patients) were found, which precluded a sound cost analysis. CONCLUSIONS: The present systematic review suggests that ERPs in colorectal and pancreas surgery are associated with cost savings compared to conventional perioperative management. Cost data in liver surgery are scarce.

Impact of ERAS compliance on the delay between surgery and adjuvant chemotherapy in hepatobiliary and pancreatic malignancies.

BACKGROUND: Multidisciplinary approach with adjuvant chemotherapy is the key element to provide optimal outcomes in pancreas and liver malignancies. However, post-operative complications may increase the interval between surgery and chemotherapy with negative oncologic effects. HYPOTHESIS AND STUDY AIM: The aim of the study was to analyse whether compliance to Enhanced Recovery After Surgery (ERAS) pathway was associated with decreased interval to adjuvant chemotherapy. METHODS: Retrospective analysis of all consecutive ERAS patients with surgery for hepatobiliary or pancreatic malignancies at the University Hospital of Lausanne between 2012 and 2016. Multivariate analysis was performed to assess the impact of ERAS compliance on time to chemotherapy. RESULTS: A total of 133 patients with adjuvant chemotherapy were included (n = 44 liver and n = 89 pancreatic cancer). Median compliance to ERAS was 61% (IQR 55-67) for the study population, and median delay to chemotherapy was 49 days (IQR 39-61). Overall, compliance ≥ 67% to ERAS induced a significant reduction in the interval between surgery and chemotherapy for young patients (< 65 years old) with or without severe comorbidities (reduction of 22 and 10 days, respectively). High compliance in young ASA3 patients with liver colorectal metastases was associated with an increase of 481 days of DFS. CONCLUSIONS: ERAS compliance ≥ 67% tends to be associated with a reduction in the delay to adjuvant chemotherapy for young patients with hepatobiliary and pancreatic malignancies. More prospective studies with strict adhesion to the ERAS protocol are needed to confirm these results.

Acute kidney injury after liver surgery: does postoperative urine output correlate with postoperative serum creatinine?

BACKGROUND: Acute kidney injury (AKI) after hepatectomy occurs in around 10% of cases. AKI is often defined based only on postoperative serum creatinine increase. This study aimed to assess if postoperative urine output (UO) correlated with serum creatinine after hepatectomy. METHODS: All consecutive hepatectomy patients (2010-2016) were assessed. AKI was defined according to KDIGO criteria: serum creatinine increase ≥26.5 µmol/l, creatinine increase ≥1.5x baseline creatinine, or postoperative oliguria. Oliguria was defined as daily mean UO <0.5 mL/kg/h. AKI was subdivided into creatinine-based or oliguria-based AKI according to the defining criterion. RESULTS: Out of 285 patients, AKI was observed in 79 cases (28%). Creatinine-based AKI occurred in 25 patients (9%) and oliguria-based only AKI in 54 patients (19%). Ten patients fulfilled both criteria (4%). Postoperative UO correlated poorly with postoperative serum creatinine level in both whole cohort (rho = -0.34, p <0.001) and AKI subgroup (rho = -0.189, p = 0.124). No association was found between postoperative oliguria and postoperative serum creatinine increase (HR = 0.5, 95%CI: 0.2-1.9, p = 0.341). On multivariable analysis, operation duration >360 minutes was the only predictor of creatinine increase (HR = 3.6, 95%CI: 1.1-11.4, p = 0.032). CONCLUSION: Postoperative UO showed poor correlation with postoperative serum creatinine both in all patients and AKI patients. Surgery duration >360 minutes appeared as the only independent predictor of postoperative serum creatinine increase.

Perioperative nutritional management of patients undergoing pancreatoduodenectomy: an international survey among surgeons.

BACKGROUND: There is still a lack of good evidence regarding the optimal perioperative nutritional management for patients undergoing pancreatoduodenectomy (PD). The aim of this international survey was to assess the current practice among pancreatic surgeons. METHODS: A web survey of 30 questions was sent to the members of the European-African Hepato-Pancreato-Biliary Association (E-AHPBA) and International Hepato-Pancreato-Biliary Association (IHPBA). All members were invited by email to answer the online survey. A reminder was sent after 4 weeks. RESULTS: In total 420 out of 2500 surgeons (17%) answered the survey. Almost half of the surgeons (44%) did not organize a preoperative nutritional consultation for their patients. Seventy-seven percent of the participants did not have specific nutritional thresholds before the operation. A majority (66%) routinely used biological parameters to detect or follow malnutrition. Regarding intraoperative details, 69% of the respondents routinely leaved a nasogastric tube at the end of PD for gastric drainage. Sixty-six percent of the participants reported a postoperative nutritional follow-up consultation during hospitalization, and 58% of them had established local standardized protocols for postoperative nutritional support. CONCLUSION: Management of perioperative nutrition in patients undergoing PD was very disparate internationally. No specific preoperative nutritional thresholds were used, and postoperative feeding routes and timing were diverse.

Impact of postoperative weight gain on complications after liver surgery.

BACKGROUND: Recent data has suggested that excessive perioperative weight gain may be associated with adverse outcomes after abdominal surgery, but this observation remains unexplored following liver surgery. The present study aimed to investigate the predictive value of perioperative weight fluctuation in predicting complications after liver surgery. METHODS: Retrospective monocentric analysis of consecutive patients undergoing liver surgery between 2010 and 2016. Patients without available perioperative weight were excluded. Test variable was postoperative weight change (ΔWeight) measured on day 2 (POD2). Primary outcome was postoperative major morbidity according to Clavien classification (grades III-IV). Secondary outcomes were overall complications, Comprehensive Complication Index (CCI) and length of hospital stay (LoS). Area under the receiver operating characteristic curve (AUROC) and logistic regression with multivariable analysis were performed. RESULTS: A total of 181 patients met the inclusion criteria. Major and overall postoperative complications were reported in 25 (14%) and 87 (48%) patients, respectively. On POD2, median ΔWeight was 2.6 Kg (IQR: 1.1-4.0). Patients with major complications showed increased ΔWeight of 4.2 Kg (IQR: 2.7-5.7), compared to 2.3 Kg (IQR: 0.9-3.7) in patients without major complications (p < 0.001). AUROC of ΔWeight for major complications was 0.74, determining an optimal cut-off of 3.5 Kg, which yielded a negative predictive value of 94%. Multivariable analysis identified ΔWeight ≥3.5 Kg as independent predictor of major complications (OR, 4.73; 95% CI, 1.51-14.80; p = 0.008). CONCLUSION: ΔWeight ≥3.5 Kg was independently associated with major complications after liver surgery. Perioperative fluctuation of weight appears as an important predictor of adverse outcomes after liver surgery.

Cost-analysis of Enhanced Recovery After Surgery (ERAS) program in gynecologic surgery.

OBJECTIVES: Enhanced recovery after surgery (ERAS) programs has shown clinical benefits in gynecologic surgery. The aim of the present study was to compare costs before and after implementation of an ERAS program for gynecologic surgery. METHODS: Retrospective study comparing perioperative costs between consecutive patient groups undergoing gynecologic surgery (benign, staging or debulking) (I, 2012-13) prior, (II) immediately after, and (III, 2014-16) the three years after ERAS implementation. Preoperative, intraoperative, and postoperative real costs were collected for each patient via hospital administration. A bootstrap independent t-test was used for comparison. RESULTS: Demographics and preoperative characteristics were similar between group I (n = 42), II (n = 51), and III (ERAS I; n = 122, II; n = 134, III; n = 90). Average ERAS-specific costs were $687 per patient. Total mean individual costs per patient were $13'329 (95% confidence interval (CI): 11'301-15'213) and $17'710 (95% CI: 14'452-21'605) in the ERAS and pre-ERAS groups respectively, resulting in net savings of $4'381 (95% CI: 549-8'752, p = 0.043) in favour of ERAS group. Cost savings were explained by lower pre- and postoperative costs (difference: $5'011 95% CI: 1'587-8'998, p = 0.019). Total costs continued to decrease by $2'520 (mean: $15'190, 95% CI: 13'791-16'631) in year 1, by $3'077 (mean: $14'633, 95% CI: 13'378-16'250) and $5'070 (mean: $12'640, 95% CI: 11'460-14'015) (p = 0.03) respectively, in year 2 and 3 after implementation. CONCLUSION: Based on real costs and including specific costs due to ERAS implementation, ERAS program in gynecologic surgery induced significant decrease of overall costs by $4'381 per patient. Total costs continued to decrease in the three years after implementation.

Tumoral Venous Invasion after Distal Pancreatectomy: A Risk Factor for Recurrence.

BACKGROUND: Few data exist on postoperative outcomes of patients with pancreatic body-tail malignancies and tumoral venous invasion (VI). This study aimed at comparing survival and recurrence rate (RR) after distal pancreatectomy for adenocarcinoma in patients with and without tumoral VI. METHODS: All consecutive distal pancreatectomies (2000-2015) were collected. Demographics and peri- and postoperative data were recorded. Survivals were calculated using Kaplan-Meier curves. RESULTS: A total of 45 patients underwent distal pancreatectomies for malignancies, of which 33 patients had ductal adenocarcinomas and 2 had cystadenocarcinomas. Among these 35 adenocarcinomas, histological VI was found in 28 patients (80%). Characteristics and intraoperative data of patients with and without VI were similar. Complication rates were 15 of 28 (54%) in the VI group and 3 of 7 (43%) in the group without VI (p = 0.612). Five-year survival for the group with and without VI were 19 and 39% (p = 0.232), respectively. RR was 16 of 28 (57%) for the VI group and 1 of 7 (14%) for the group without VI (p = 0.042). CONCLUSION: VI did not have an effect on postoperative -complications. Survivals were similar in case of VI or not. On the contrary, RR was higher in the VI group.

Beyond surgery: clinical and economic impact of Enhanced Recovery After Surgery programs.

BACKGROUND: Enhanced Recovery After Surgery (ERAS) is a perioperative management based on multimodality and multidisciplinary work. ERAS has been shown to have important clinical and economic benefits, but its spread remains slow worldwide. DISCUSSION: This manuscript reviews the overall program benefits and focuses on important aspects for implementation well beyond surgery. Implementation of ERAS pathways improves clinical outcomes and induces substantial economic gains. ERAS is the current surgical revolution.

Antibiotic treatment for uncomplicated and mild complicated diverticulitis: outpatient treatment for everyone.

PURPOSE: Antibiotic treatment is the treatment of choice for uncomplicated diverticulitis (uD) and can be performed for mild complicated diverticulitis (mcD). In several cases, outpatient treatment (OT) can be undertaken. This study assessed the 1-month failure rate of OT for uD/mcD compared to inpatient treatment (IT), and identified predictive factors for treatment failure. METHODS: All consecutive patients (2006-2012) diagnosed with uD/mcD by CT scan were retrospectively analyzed. Acute uD was defined as absence of the following: abscess, fistula, extraluminal contrast, pneumoperitoneum, and need for immediate percutaneous drainage/surgery. Acute mcD was defined as complicated diverticulitis with abscess <4 cm or pneumoperitoneum <2 cm. All patients received antibiotherapy. Treatment failure was defined as (re)hospitalization the first month after treatment onset or need of drainage/surgery during hospitalization. All patients were contacted using a standardized questionnaire. RESULTS: Out of 540 uD/mcD, IT was offered to 369 patients (68%) and OT to 171 patients (32%). The IT group had higher median age, more women, higher median Charlson Index, more severe median Ambrosetti score, longer median time in the emergency room, and higher median CRP. Response rates to the questionnaire were 56% (IT) vs. 62% (OT), p = 0.18. Failure rates were 32% in IT vs. 10% in OT group, p < 0.01. Among the uD/mcD patients, admission/CT time between midnight and 6 AM, Ambrosetti score of 4, and free air around the colon were risk factors for failure. CONCLUSIONS: Outpatient treatment for uncomplicated/mild complicated diverticulitis is feasible and safe. Prognostic factors of failure necessitating closer follow-up were admission/CT time, Ambrosetti score of 4, and free air around the colon.

Treatment and outcomes of recurrent hepatocellular carcinomas.

PURPOSE: Surgery is one of the best options for curative treatment of hepatocellular carcinomas (HCC). Recurrences are nevertheless common (45-75%). This study aimed to compare overall survival (OS) of patients with recurrent HCC after primary resection to OS of patients without recurrence. METHODS: A retrospective review of all HCC patients operated between 1993 and 2015 was performed. Median and 5-year OS were calculated. RESULTS: This study included 147 HCC patients. Sixty-seven patients presented a recurrence (46%). Patients with recurrence had a worse prognosis than those without recurrence (median OS 63 vs. 82 months, 5-year OS 47 vs. 54%, p = 0.036). First-line performed treatments were radiofrequency ablation (18, RFA), chemo-embolization (16, TACE), repeat hepatectomy (10), systemic chemotherapy (4), radio-embolization (1), and alcoholization (1). Palliative care was performed in 17 patients. Median OS of patients treated by RFA, TACE, or repeat hepatectomy were similar (77, 71, and 84 months, p = 0.735). Patients treated with chemotherapy/palliative care had lower median OS compared to interventional treatments (20 vs. 77 months, p < 0.0001). CONCLUSIONS: Recurrence after surgical HCC resection is frequent and negatively impacts OS. Interventional treatments of recurrences offered improved outcomes compared to medical care. In selected patients, RFA, TACE, and repeat hepatectomy allowed similar OS as non-recurrent cases.