Damien Bergeat

Gastric and oesophageal cancers are common. They are also expected to increase in incidence in the next few years and are characterized by poor prognosis. Surprisingly, whereas the incidence of severe anxiety and depression is high in patients with gastric and oesophageal cancers, the influence of symptoms of depression and anxiety on postoperative complications has barely been explored. In a retrospective study based on a prospectively collected database, 629 cancer patients were enrolled. Symptoms of depression and anxiety (Hospital Anxiety and Depression Scale scores) and sociodemographic and medical information were collected immediately after diagnosis and before any treatment. The surgical approach (i.e. gastrectomy or oesophagectomy) and postoperative complications according to the Clavien-Dindo classification were collected after surgery. After controlling for known medical predictors (i.e. surgical strategy, alcohol and tobacco consumption, American Society of Anaesthesiologists classification physical status score) of postoperative complications, no effect of symptoms of depression or anxiety was detected. The observed results are surprising given the literature. However, several potential arguments can be put forwards regarding methods and measures, controlling variables, and conceptual distinctions. Despite the absence of significant results, this topic should be more deeply investigated by applying methodological and conceptual adjustments.

The outcomes of bariatric surgery (BS) in patients with chronic inflammatory bowel disease (IBD) remain rarely described. We aimed to evaluate the 90-day morbidity and mortality rates, and the risk of IBD complications 2 years after BS. Patients from the French Programme de Médicalisation des Systèmes d’Information (PMSI) database who underwent a primary BS between 2016 and 2018 were included. We identified patients with a previous diagnosis of IBD. Postoperative 90-day (POD90) morbidity and mortality rates were compared between the two groups. The evolution of IBD was followed 2 years after BS. Between 2016 and 2018, 138 980 patients underwent primary BS, including 587 patients with IBD: 326 (55.5 per cent) with Crohn’s disease (CD) and 261 (44.5 per cent) with ulcerative colitis (UC). The preferred surgical technique was sleeve gastrectomy, especially in the IBD group (81.1 per cent), followed by gastric bypass (14.6 per cent). Patients with IBD had more comorbidities (Charlson Comorbidity Index of 1 or more, hypertension, and diabetes; P < 0.001) than those without IBD. The POD90 mortality rate did not differ between the two groups (0.049 per cent in the IBD group versus 0 per cent in the non-IBD group), but more unscheduled rehospitalizations at POD90 were observed in patients with IBD (6.0 per cent versus 3.7 per cent; P = 0.004). Two years after BS, 86 patients (14.6 per cent) in the IBD group had at least one unplanned readmission for the management of their IBD; 15 patients stayed for 3 or more days. After multivariable analysis, patients with CD had an independent elevated risk of IBD-related unplanned readmissions 2 years after BS versus UC (adjusted odds ratio 1.90, 95 per cent c.i. 1.22 to 2.97; P = 0.005). In a highly selected cohort of patients with well-controlled IBD, BS did not result in added mortality or morbidity. A point of vigilance must be underlined regarding BS in patients with CD.

The impact of body composition (BC) on the prognosis of resected intrahepatic cholangiocarcinoma (ICC) has been poorly studied. Aims: i) to evaluate the prevalence of low muscle mass (MM) in patients; ii) to assess the impact of BC on patient overall survival (OS) and disease-free survival (DFS), and iii) on the incidence of postoperative complications. All consecutive patients who underwent liver resection for ICC between 2004 and 2016 and who had preoperative CT scans were included. Ninety-three patients were included. Sixty percent (55/91) had low total MM. On multivariable analysis, high visceral fat (HR 2.48, CI95% [1.63; 3.77], p < 0.0001), nodules >1 (HR 3.15 [1.67; 5.93], p = 0.0004), involvement adjacent organ (HR 6.67 [1.88; 23.69], p = 0.003), and postoperative sepsis (HR 3.04 [1.54; 5.99], p = 0.0013) were independently associated with OS. High visceral fat (HR 2.10 [1.31; 3.38], p = 0.002], nodules >1 (HR 3.01, [1.49; 6.10], p = 0.002), postoperative sepsis (HR 5.16 [2.24; 11.89], p = 0.0001), ASA score (p = 0.02) and perineural invasion (HR 3.30 [1.62; 6.76], p = 0.001) were independently associated with lower DFS. Conclusion: 60% of ICC patients had low MM before surgery. High visceral fat, but not muscle mass, was an independent prognostic factor for poor OS and DFS in European patients with resected ICC.

Objective: The aim of the study was to evaluate the impact of the use of a reinforced stapler (RS) during distal pancreatectomy (DP) on postoperative outcomes. Background: DP remains associated with significant postoperative morbidity owing to pancreatic fistula (PF). To date, there is no consensus on the management of the pancreatic stump. The use of an RS potentially represents a simple way to decrease the rate of PF. Methods: The REPLAY study (NCT03030170) is a prospective, multicenter, randomized study. Patients who underwent DP were randomized (1:1 ratio) in 2 groups for the use of a standard stapler (SS) or an RS to close remnant pancreatic parenchyma. The primary endpoint was the rate of overall PF. Secondary endpoints included severity of PF, length of hospital stay, overall morbidity, and rate of readmission for a PF within 90 days. Participants were blinded to the procedure actually carried out. Results: A total of 199 were analyzed (SS, n=99; RS, n=100). One patient who did not undergo surgery was excluded. Baseline characteristics were comparable in both groups. The rate of overall PF was higher in RS group (SS: 67.7%, RS: 83%, P=0.0121), but the rate of clinically relevant PF was similar (SS: 11.1%, RS: 14%, P=0.5387). Mean length of total hospital stay, readmission for PF, postoperative morbidity, and mortality at 90 days were similar. Conclusion: The results of this randomized clinical trial did not favor the use of RS during DP to reduce the rate of PF.

How the side of an extended liver resection impacts the postoperative prognosis of advanced perihilar cholangiocarcinoma (PHC) is still controversial. We compared the outcomes of right (RTS) and left trisectionectomies (LTS) in Bismuth-Corlette (BC) type IV PHC resection. All patients undergoing RTS or LTS for BC type IV PHC in a single tertiary center between January 2012 and December 2019 were compared retrospectively. The endpoints were perioperative outcomes, long-term overall (OS), and disease-free survival (DFS). Among 67 hepatic resections for BC type IV PHC, 25 (37.3%) were LTS and 42 (63.7%) were RTS. Portal vein and artery resection rates were 40% and 52.4% (p = 0.29), and 24% and 0% (p < 0.001) in the LTS and RTS groups, respectively. The severe complication (Clavien–Dindo > IIIa) rate was comparable (36% vs. 21.5%, p = 0.357) while the postoperative liver failure (POLF) rate was lower in the LTS group (16% vs. 38%, p = 0.048). The R0 resection rate was similar between groups (81% vs. 92%; p = 0.154). The five-year OS rate was higher in the LTS group (66% vs. 30%, p = 0.009) while DFS was comparable (43% vs. 18%, p = 0.11). Based on multivariable analysis, the side of the trisectionectomy was an independent predictor of OS. Compared with RTS, LTS is associated with lower POLF and higher overall survival despite more frequent arterial reconstructions in type IV PHC. Although technically more demanding, LTS may be preferred in the treatment of advanced PHC.

Randomized clinical trials (RCTs) are the building blocks of an evidence-based approach to medicine and surgery. These trials are used as trusted sources of evidence when designing clinical practice guidelines. It is therefore expected that their results are trustworthy and that their conclusions can be verified. Randomized clinical trial data sharing (ie, sharing statistical codes, study protocols, and individual participant data) holds promise for enhancing transparency and maximizing the value of research.¹ Data sharing is supported by patients who also want to share their data in a responsible and secure manner.² In 2018, the International Committee of Medical Journal Editors (ICMJE) required that manuscripts submitted to ICMJE journals as of July 1, 2018, must contain a data sharing statement if they are reporting the results of an RCT. If a trial began enrolling participants on or after January 1, 2019, the article must also include a data sharing plan in the trial’s registration.³ The ICMJE considers that “it is an ethical obligation to responsibly share data generated by interventional clinical trials because participants have put themselves at risk.”⁴