- Email: [email protected]
Press review
Journal of Visceral Surgery (2016) 153, 465—472
Available online at
ScienceDirect www.sciencedirect.com
Press review C. Mariette a,∗, S. Benoist b a
Service de chirurgie digestive et générale, hôpital C.-Huriez, CHRU, place de Verdun, 59037 Lille cedex, France b Service de chirurgie digestive, hôpital du Kremlin-Bicêtre, 78, rue du Général-Leclerc, 94275 Le Kremlin-Bicêtre, France Available online 16 November 2016 Cousin F, Ortega-Deballon P, Bourredjem A, et al. Diagnostic accuracy of procalcitonin and C-reactive protein for the early diagnosis of intra-abdominal infection after elective colorectal surgery: a meta-analysis. Ann Surg 2016;264:252—6. http://dx.doi.org/10.1097/SLA. 0000000000001545
Objective Intra-abdominal infections (IAIs) after elective colorectal surgery impact significantly the short- and long-term outcomes. In the era of fast-track surgery, they often come to light after discharge from hospital. Early diagnosis is therefore essential. C-reactive protein levels have proved to be accurate in this setting. Procalcitonin has been evaluated in several studies with conflicting results. This meta-analysis aimed to compare the predictive abilities of C-reactive protein and procalcitonin in the occurrence of IAIs after elective colorectal surgery. Methods This meta-analysis included studies analyzing C-reactive protein and/or procalcitonin levels at postoperative days 2, 3, 4, and/or 5 as markers of intra-abdominal infection after elective colorectal surgery. Methodological quality was assessed by the QUADAS2 tool. The area under the curve summary receiver-operating characteristic was calculated for each day and each biomarker, using a random-effects model in cases of heterogeneity. Results The meta-analysis included 11 studies (2692 patients). An IAI occurred in 8.9% of the patients. On postoperative day 3, area under the curve was 0.80 (95% CI, 0.76—0.85) for C-reactive protein and 0.78 (95% CI, 0.68—0.87) for procalcitonin. On postoperative day 5, their predictive accuracies were 0.87 (95% CI, 0.80—0.93) and 0.90 (95% CI, 0.82—0.98),
∗
Corresponding author. E-mail address: [email protected] (C. Mariette).
1878-7886/$ — see front matter http://dx.doi.org/10.1016/j.jviscsurg.2016.09.018
respectively. The accuracy of C-reactive protein and procalcitonin did not differ at any postoperative day. Conclusions Levels of inflammatory markers under the cutoff value between postoperative days 3 and 5 ensure safe early discharge after elective colorectal surgery. Procalcitonin seems not to have added value as compared to C-reactive protein in this setting. Comments 1. This meta-analysis confirms the results of the French prospective study organized by the same authors that showed that procalcitonin (PCT) was not useful for the early detection of infectious complications after colorectal surgery [1]. 2. This meta-analysis, however, has several methodological biases. Firstly, the heterogeneity in the definition of the main end-point, i.e. IAI. Some of the studies only included anastomotic leaks, while others also included simple intra-abdominal collections seen on imaging. Moreover, in all the studies, the manner to search for IAI was not the same, and therefore, some might have escaped diagnosis. Last, in all the studies except one, the investigators whose role was to look for IAI postoperatively were aware of the values of CRP and PCT, which might have encouraged the investigators to look for IAI in asymptomatic patients with high values, thus increasing the positive and negative predictive values of these two investigations. 3. The main question for which this meta-analysis does not provide any answer is to know whether early detection of IAI in pauci-symptomatic patients can change ulterior management. Reference [1] Ann Surg 2016;263:967—72. Yamada T, Okabayashi K, Hasegawa H, et al. Meta-analysis of the risk of small bowel obstruction following open or laparoscopic colorectal surgery. Br J Surg 2016;103:493—503. http://dx.doi.org/10.1002/bjs.10105
466
C. Mariette, S. Benoist
Background One of the potential advantages of laparoscopic compared with open colorectal surgery is a reduction in postoperative bowel obstruction events. Early reports support this proposal, but accumulated evidence is lacking. Methods A systematic review and meta-analysis was performed of randomized clinical trials and observational studies by searching the PubMed and Cochrane Library databases from 1990 to August 2015. The primary outcomes were early and late postoperative bowel obstruction following laparoscopic and open colorectal surgery. Both ileus and bowel obstruction were defined as a postoperative bowel obstruction. Subgroup and sensitivity analyses were performed, and a random-effects model was used to account for the heterogeneity among the studies. Results Twenty-four randomized clinical trials and 88 observational studies were included in the meta-analysis; 106 studies reported early outcome and 12 late outcome. Collectively, these studies reported on the outcomes of 148—392 patients, of whom 58—133 had laparoscopic surgery and 90—259 open surgery. Compared with open surgery, laparoscopic surgery was associated with reduced rates of early (odds ratio 0.62, 95 per cent c.i. 0.54 to 0.72; P < 0.001) and late (odds ratio 0.61, 0.41 to 0.92; P = 0.019) postoperative bowel obstruction. Weighted mean values for early postoperative bowel obstruction were 8 (95 per cent c.i. 6 to 10) and 5 (3 to 7) per cent for open and laparoscopic surgery respectively, and for late bowel obstruction were 4 (2 to 6) and 3 (1 to 5) per cent respectively. Conclusion The reduction in postoperative bowel obstruction demonstrates an advantage of laparoscopic surgery in patients with colorectal disease. Comments 1. In patients with early bowel obstruction, the authors did not distinguish between post-operative ileus and true obstruction due to adhesions or bands. Moreover, there were no precisions given concerning the diagnosis of obstruction: was it based on clinical, surgical or radiological findings? 2. With respect to late bowel obstruction, there was no information concerning the mean follow-up in the different studies. It is likely that this period was short, as the 4% post-laparotomy rate appears to be extremely low. A longer follow-up period would have permitted inclusion of more events, and therefore a more precise evaluation of the magnitude of the effect. 3. While this meta-analysis is not perfect methodologically speaking, it constitutes another argument in favor of the laparoscopic approach in colorectal surgery.
Bains SJ, Mahic M, Myklebust TÅ, et al. Aspirin as secondary prevention in patients with colorectal cancer: an unselected population-based study. J Clin Oncol 2016;34:2501—2508. http://dx.doi.org/10.1200/JCO.2015. 65.3519
Purpose Regular use of aspirin (acetylsalicylic acid) is associated with reduced incidence and mortality of colorectal cancer (CRC). However, aspirin as primary prevention is debated because of the risk of hemorrhagic adverse effects. Aspirin as secondary prevention may be more justified from
a risk-benefit perspective. We have examined the association between aspirin use after the diagnosis of CRC with CRC-specific survival (CSS) and overall survival (OS). Materials and methods An observational, population-based, retrospective cohort study was conducted by linking patients diagnosed with CRC from 2004 through 2011 (Cancer Registry of Norway) with data on their aspirin use (The Norwegian Prescription Database). These registries cover more than 99% of the Norwegian population and include all patients in an unselected and consecutive manner. Exposure to aspirin was defined as receipt of aspirin prescriptions for more than 6 months after the diagnosis of CRC. Multivariable Cox-proportional hazard analyses were used to model survival. The main outcome measures of the study were CSS and OS. Results A total of 23,162 patients diagnosed with CRC were included, 6102 of whom were exposed to aspirin after the diagnosis of CRC (26.3%). The median follow-up time was 3.0 years. A total of 2071 deaths (32.9%, all causes) occurred among aspirin-exposed patients, of which 1158 (19.0%) were CRC specific. Among unexposed patients (n = 17,060), there were 7218 deaths (42.3%), of which 5375 (31.5%) were CRC specific. In multivariable analysis, aspirin exposure after the diagnosis of CRC was independently associated with improved CSS (hazard ratio [HR], 0.85; 95% CI, 0.79 to 0.92) and OS (HR, 0.95; 95% CI, 0.90 to 1.01). Conclusion Aspirin use after the diagnosis of CRC is independently associated with improved CSS and OS. Comments 1. The interpretation of the conclusions warrants caution because of the many methodological biases in this study. Firstly, the patients in each group were not comparable, particularly for the major prognostic factors such as disease stage or tumor differentiation. Effectively, tumors in the aspirin group were less locally advanced and more differentiated than in the control group. Moreover, in addition to surgery, there was no information concerning the treatment for cancer and in particular, concerning chemotherapy. Lastly, several prognostic factors such as CEA, the R0 resection rate, administration of chemoradiation in patients with rectal cancer, the site of synchronous metastases if present, were not included in the multivariable analysis. 2. From the methodological point of view, since the two groups were not absolutely comparable, the authors should have performed a propensity score using logistic regression including the variables that differed between the two groups, and then either include this score in the multivariable model, or match the patients using the score. 3. It is difficult to determine whether aspirin use after the diagnosis of CRC improved cancer-specific survival, as it was in patients who took aspirin before and after diagnosis that the greater benefit was observed. It is also probable that there was a greater benefit for patients who took aspirin before diagnosis (vs. after) since cancerspecific survival was not improved in the subgroup of patients who only took aspirin after diagnosis was made.
Fazi M, Giudici F, Luceri C, et al. Long-term results and recurrence-related risk factors for Crohn disease in patients undergoing side-to-side isoperistaltic strictureplasty. JAMA
Press review Surg 2016;151:452—60. jamasurg.2015.4552
467 http://dx.doi.org/10.1001/
Importance Side-to-side isoperistaltic strictureplasty (SSIS) is useful in patients undergoing surgery for Crohn disease (CD) to avoid wide small-bowel resections. To our knowledge, there are no definitive data regarding its recurrence risk factors. Objective To evaluate the results obtained in a monocentric population of patients with CD who have undergone SSIS. Design, setting, and participants From August 1996 to March 2010, 91 patients with CD underwent SSIS in our center. In this prospective observational study, side-to-side isoperistaltic strictureplasty was according the Michelassi technique in 69 patients and the Tonelli technique in 22 patients. Factors relating to the patient and the CD, surgery, and pharmacological therapy during the preoperative and perioperative periods were evaluated in association with medical or surgical recurrence. Exposure Side-to-side isoperistaltic strictureplasty. Main outcomes and measures The recurrence-free curve was estimated using KaplanMeier analysis. Patients were stratified into cohorts in relation to the considered categorical variables and data were compared by using the Mantel-Cox log-rank test. Cox proportional hazard regression analysis was used to set up a predictive model simultaneously exploring the effects of all independent variables on a dichotomous outcome recurrence in relation to time. Results Among the 91 patients, the mean (SD) age was 39.5 (11.2) years and preoperative disease duration was 97.9 (85.8) months; 83 patients (91.2%) were followed up, of whom 37 (44.58%) experienced a recurrence at a mean (SD) of 55.46 (36.79) months after surgery (range, 9—140 months). The recurrence in the SSIS site at a mean (SD) of 48.25 (29.94) months after surgery affected 24 of 83 patients (28.9%), 9 being medical and 15 being surgical recurrence. Recurrence in the SSIS was statistically significantly associated with the time elapsed between diagnosis and surgery (P = .03). A borderline association between family history of CD and surgical recurrence (P = .054) was also found. Multivariate analysis identified the age at diagnosis (2 = 5.56; P = .02) and at surgery (2 = 7.77; P = .005), family history (2 = 6.26; P = .01), and smoking habit (2 = 10.06; P = .007) as independent risk factors for recurrence. Conclusions and relevance In the short-term, SSIS leads to a resolution of symptoms in more than 90% of cases and the recurrence rate in the SSIS area is acceptable, even after long-term follow-up. Comments 1. This study confirms that there are very few indications for strictureplasty; in this specialized center for CD, only 91 operations with strictureplasty were performed over 12 years, i.e. fewer than eight per year. It would therefore have been interesting to know the total amount of patients undergoing operation for CD during the same period. 2. All the patients included in this series had severe disease with multiple site involvement and more than half had already undergone surgery for their CD, a well-known risk factor for recurrence after surgery [1]. Only 1/3 of these patients had received immuno-suppressive or antiTNF treatment before surgery. This means some of the
patients in this series received sub-optimal treatment since these patients would not have undergone operation today or, in any case, only after failure of several lines of therapy. Likewise, giving only Pentasa® for patients at high risk of recurrence does not seem to be appropriate, since today, most of these patients would have immuno-suppressive therapy [2]. 3. Finally, the clinical impact of this narrative study is practically nil; the results of this study will probably not lead to increased indications for long side-to-side strictureplasties; they only confirm those of a multicenter study published in 2007 [3]. Moreover, the 45% recurrence rate is comparable to that reported after resection [4]. Lastly, the risk factors found here are already well-known risk factors for recurrence. With the emergence of new treatments, the risk of short bowel syndrome is rare and the situations where strictureplasty should be preferred over resection are not common. References [1] Dis Colon Rectum 2009;52:919—27. [2] Am J Gastroenterol 2010;105:1158—64. [3] Dis Colon Rectum 2007;50:277—84. [4] Inflam Bowel Dis 2013;19:7—14. Sadot E, Lee SY, Sofocleous CT, et al. Hepatic resection or ablation for isolated breast cancer liver metastasis: a case-control study with comparison to medically treated patients. Ann Surg 2016;264:147—54. http://dx.doi.org/ 10.1097/SLA.0000000000001371
Objective To evaluate the efficacy of surgical treatment for patients with isolated breast cancer liver metastases (BCLM). Background Single-arm retrospective studies have shown promising results associated with surgery for isolated BCLM; however, this treatment remains controversial and its role is not welldefined. Methods A review of 2150 patients with BCLM who underwent treatment in a single institution was conducted, and 167 (8%) patients with isolated BCLM were identified. A case-control study was conducted to compare outcomes in patients with isolated BCLM who underwent surgery and/or ablation to patients who underwent conventional medical therapy. Results A total of 167 patients were included (surgery/ablation: 69; medical: 98), with a median follow-up for survivors of 73 months. Patients in the surgical cohort more frequently had estrogen receptor-positive tumors and received adjuvant chemotherapy and radiotherapy for their primary breast tumor. The hepatic tumor burden was less and the interval from breast cancer diagnosis to BCLM was significantly longer (53 vs 30 months) in the surgical cohort. Patients undergoing surgical treatment had a median recurrence-free interval of 28.5 months (95% confidence interval (CI): 19—38) with 10 patients (15%) recurrence free after 5 years. There was no significant difference in overall survival (OS) between the surgical and medical cohorts (median OS: 50 vs 45 months; 5-year OS: 38% vs 39%). Conclusions Hepatic resection and/or ablation was not associated with a survival advantage. However, significant recurrencefree intervals can be accomplished with surgical treatment. Surgical intervention might be considered in highly selected
468
C. Mariette, S. Benoist
patients with the goal of providing time off of systemic chemotherapy. Comments 1. This study rekindles the debate on the management of patients with BCLM and is in contradiction with the results of initial studies that were strongly in favor of surgery. Notwithstanding, the outcome of surgery here is rather good; the 40% 5-year survival rate is comparable to that reported in other surgical series [1—3]. 2. However, one could be led to think that surgery is not as good as medical treatment since survival was not improved in the surgical group even though patients in this group had less aggressive disease. A recent metaanalysis including 1341 patients with metastatic breast cancer showed that prognosis improved over the 20-year period of inclusion, a period during which surgery was performed less and less often [4]. This suggests that improved prognosis was due to new medications rather than surgery. 3. In previous studies [1—3], it was shown that the best indications for surgery were good control of BCLM by induction chemotherapy. In this study, however, patients undergoing surgery for their BCLM apparently did not receive pre-operative chemotherapy, which could have worsened the results for surgery. 4. Regrettably, the authors did not evaluate quality of life. Since patients in the surgery group had an interval median follow-up period of 25 months without chemotherapy (compared to patients in the chemotherapy group who were treated sans interruption), which could have greatly impacted quality of life. Therefore, surgery might still be the best option for BCLM, since it is associated with better quality of life without decreasing survival. References [1] Ann Surg Oncol 2004;11:869—74. [2] Ann Surg 2006;244:897—907. [3] Eur J Cancer 2011;47:2282—90. [4] Breast Cancer Res Treat 2010;119:621—31. Hof J, Wertenbroek MW, Peeters PM, et al. Prospective outcomes after resection and/or radiofrequency ablation for recurrence after treatment of colorectal liver metastases. Br J Surg 2016;103:1055—62. http://dx.doi.org/10.1002/ bjs.10162
Background Repeat liver resection for colorectal liver metastases (CRLMs) is possible in a limited number of patients, with radiofrequency ablation (RFA) as an alternative for unresectable CRLMs. The aim of this study was to analyse survival rates with these interventions. Methods This was a database analysis of patients who underwent first and repeat interventions for synchronous and metachronous CRLMs between 2000 and 2013. Descriptive and survival statistics were calculated. Results Among 431 patients who underwent resection or RFA for CRLMs, 305 patients developed recurrences for which 160 repeat interventions (resection and/or RFA or ablative radiotherapy) were performed. In total, after 707 first or repeat interventions, 516 recurrences (73.0 per cent) developed, of which 276 were retreated curatively. At the time of first intervention, independent risk factors for death were
lymph node-positive primary tumour (hazard ratio [HR] 1.40; P = 0.030), more than one CRLM (HR 1.53; P = 0.007), carcinoembryonic antigen level exceeding 200 ng/ml (HR 1.89; P = 0.020) and size of largest CRLM greater than 5 cm (HR 1.54; P = 0.014). The 5-year overall survival rates for liver resection and percutaneous RFA as first intervention were 51.9 and 53 per cent, with a median overall survival of 65.0 (95 per cent c.i. 47.3 to 82.6) and 62.1 (52.2 to 72.1) months, respectively. Conclusion RFA had good oncological outcomes in patients with unresectable CRLMs. Radiofrequency ablation is progressively more applied with each additional intervention. Comments 1. The only strong point in this series is that analysis was conducted with intention to treat. Of particular interest, iterative intervention was possible in only 53% of patients with recurrent CRLM in this unselected population. 2. It is difficult to conclude that iterative surgery increased overall survival since there was no control group treated with chemotherapy alone. Moreover, oncological results in this series were good but not exceptional, since a 48% 5-year survival rate is comparable to those reported in other series where iterative interventions were less frequent [1,2]. 3. The main message of this study is to show that radiofrequency ablation allows iterative interventions. Effectively, radiofrequency ablation alone was used firstline in 30% of patients and in 50% beginning with the second re-intervention. References [1] Lancet Oncol 2013;14:1208—15. [2] J Clin Oncol 2007;25:4575—80. Fang X, Baillargeon JG, Jupiter DC. Continued antiplatelet therapy and risk of bleeding in gastrointestinal procedures: a systematic review. J Am Coll Surg 2016;222:890—905. http://dx.doi.org/10.1016/j.jamcollsurg.2016.01.053
Background Management of peri-operative antiplatelet medications in gastrointestinal (GI) surgery is challenging. The risk of intraoperative and postoperative bleeding is associated with perioperative use of antiplatelet medication. However, cessation of these drugs may be unsafe for patients who are required to maintain antiplatelet use due to cardiovascular conditions. The objective of this systematic review was to compare the risk of intraoperative or postoperative bleeding among patients who had GI surgery while on continuous antiplatelet therapy (aspirin, clopidogrel, or dual therapy) with the risk among those not taking continuous antiplatelet medication. Study design We reviewed articles published between January 2000 and July 2015 from the Medline Ovid and Cumulative Index to Nursing and Allied Health Literature (CINAHL) databases. Studies involving any GI procedures were included if the articles met our inclusion criteria (listed in Methods). The following keywords were used for the search: clopidogrel, Plavix, aspirin, antiplatelet, bleeding, hemorrhage, and digestive system surgical procedures. Quality of the studies was assessed, depending on their study design, using the Newcastle-Ottawa score or the Cochrane Collaboration’s tool for assessing risk of bias.
Press review Results Twenty-two studies were eligible for inclusion in the systematic review. Five showed that the risk of intraoperative bleeding or postoperative bleeding among patients who had GI surgery while on continuous antiplatelet therapy was higher compared that for those not on continuous therapy. The remaining 17 studies reported that there was no statistically significant difference in the risks of bleeding between the continuous antiplatelet therapy group and the group without continuous antiplatelet therapy. Conclusions The risk of bleeding associated with GI procedures in patients receiving antiplatelet therapy was not significantly higher than in patients with no antiplatelet or interrupted antiplatelet therapy. Comments 1. The highlight of this study is that it goes against the dogma recommending that antiplatelet therapy (APT), and particularly, aspirin, should be discontinued before surgery or gastro-intestinal endoscopy, an idea that has prevailed for decades. The risk-benefit ratio favors continuation of APT before, during and after these procedures. 2. Regrettably, no meta-analysis was performed; a statistical compilation of events would have provided a more precise idea of the risk, evaluated by an odds ratio, whether for hemorrhage in the group with APT, or ischemia in the group with discontinuation of APT. 3. It might have been interesting to have a precise evaluation of the cardiovascular risk in case of discontinued APT. 4. The level of evidence is high for aspirin but much lower for clopidrogel as there were only four studies dedicated to this APT. Consequently, while waiting for further studies, it seems reasonable to follow the recommendations laid down by the American Society for Gastrointestinal Endoscopy (ASGE) [1]: continue aspirin, but discontinue clopidrogel 7 to 10 days before gastro-intestinal surgery in patients with a low thrombo-embolic risk and weigh the risk-benefit ratio for patients at high risk. 5. The limitations of this study include the lack of information about: the doses used, the period of exposure before surgery, identification of subgroups for which this strategy should be modeled as well as quantitative analysis of results. Reference [1] Gastrointest Endosc 2009;70:1060—70. Barazanchi AW, Fagan PV, Smith BB, Hill AG. Routine neurectomy of inguinal nerves during open onlay mesh hernia repair: a meta-analysis of randomized trials. Ann Surg 2016;264:64—72. http://dx.doi.org/10.1097/SLA. 0000000000001613
Objective The aim of the study was to establish whether an inguinal neurectomy at the time of hernia repair would reduce the risk of postoperative pain for open tension-free sutured mesh repair. Background Inguinal hernia repair is a common operative procedure. The development of postoperative pain is uncommon, but at times debilitating. The role of inguinal neurectomy is currently unknown, with no single large study available, and previous reviews included only a few heterogeneous studies.
469 Methods Relevant randomized trials were identified from searches of MEDLINE, EMBASE, and EBM Review databases until October 2014. Meta-analysis was performed based on Cochrane Methods using RevMan v5.3 software. Pain, pain scores, sensory changes, and complications over short (half to < 3 months), mid (3 to < 12 months), and long term (≥ 12 months) were recorded. Results All included studies performed Lichtenstein hernia repair. Eleven studies on 1031 patients showed significant reduction in pain with neurectomy for short (RR = 0.61, 0.40-0.93) and midterm (RR = 0.30, 0.20—0.46), but not for long term (RR = 0.50, 0.25—1.01). Three studies (270 patients) showed significantly reduced short-term pain (RR = 0.69, 0.52—0.90). No studies included genitofemoral neurectomy. Rates of hematoma, infection, urinary retention, and recurrence were not different between groups. Conclusions Routine ilioinguinal neurectomy during Lichtenstein-type herniorrhaphy seems to be a safe and effective method to reduce pain in the short and midterm, but may have little long-term impact. Iliohypogastric neurectomy seems to reduce pain in at least the short term. Comments 1. The Lichtenstein open repair is the standard inguinal hernia repair technique for the American College of Surgeons [1], having been shown to be the safest of inguinal hernia repair techniques [2]. The European Hernia Society proposes laparoscopic repair but the learning curve is long [3]. 2. While recurrence is low, the principal complication reported is chronic post-operative pain [4], iatrogenic nerve injury being the most frequently reported promoting factor [5]. 3. There is no explanation for increased pain when nerves are spared. However, inflammation, entrapment or perineural fibrosis have been incriminated. The role of prosthetic material is not well known. References [1] Hernia 2004;8:1—7. [2] Ann Surg 2012;255:784—88. [3] Hernia 2014;18:151—63. [4] Ann Surg 2006;244:212—9. [5] Lancet 2006;367:1618—25. Kadry Z, Schaefer EW, Shah RA, et al. Portal hypertension: an underestimated entity? Ann Surg 2016;263:986—91. http://dx.doi.org/10.1097/SLA.0000000000001299
Objective The aim of this study is to evaluate portal hypertension as an independent risk factor in general surgical procedures. Background Data on the impact of portal hypertension in general surgical outcomes has been limited. Published literature has focused mainly on its effect in liver surgery. The Child Pugh score and Model for End Stage Liver Disease are utilized for surgical risk assessment in liver disease but they do not accurately reflect degree of portal hypertension. Methods From 2005 to 2012, patients with esophageal varices (EV) in the National Surgical Quality Improvement Program (NSQIP) formed the portal hypertension cohort, and were case matched to patients without esophageal varices (NEV)
470
C. Mariette, S. Benoist
based on sex, age, surgery type, and year of operation. Thirty day mortality and morbidity were analyzed using generalized estimating equations for binary outcomes. EV patients were also dichotomized by Model for End Stage Liver Disease (MELD) score (≤15 vs >15) and compared with NEV patients. Results One thousand five hundred and seventy-four EV patients were matched to 3148 NEV patients. In multivariable analysis, EV patients had a 3.01 higher odds of 30 day mortality (P < 0.001) and 1.28 higher odds of complications (P < 0.001) compared with NEV patients. EV patients with MELD >15 had 4.64 higher odds of death within 30 days (P < 0.001) and had 1.75 higher odds of complications within 30 days (P < 0.001) compared with NEV patients; EV patients with MELD 15 or less had 1.95 higher odds of 30 day mortality (P < 0.001) compared with NEV patients. Conclusions Portal hypertension is associated with a significant mortality and morbidity risk in general surgery, and should not be underestimated even in patients with MELD 15 or less where the early mortality risk remained significant. Comments 1. Portal hypertension (PHT) is associated with increased mortality and morbidity in liver surgery, particularly in patients with cirrhosis [1,2]. Here the authors underscored that PHT significantly increased post-operative mortality and morbidity after general, non-hepatic surgery. 2. The value of Child-Pugh’s score is limited by the subjectivity in the evaluation of certain variables entering the score such as the degree of ascites and encephalopathy. The MELD score, conversely, does not explore any of the components of PHT and is limited to three variables: total bilirubinemia, international normalized ratio (INR) and creatininemia, which can be affected by non-liver causes. The results of this study were that the MELD score influences the risk of post-operative death in patients with EV, even in patients where the MELD score ≤15. 3. Regrettably, PHT was evaluated only by the presence of EV; some patients could have had PHT without EV and therefore might have been included in the NEV group. References [1] Gastroenterology 1996;111:1018—22. [2] Ann Surg 2006;243:373—9. Brown SR, Tiernan JP, Watson AJ, et al. Haemorrhoidal artery ligation versus rubber band ligation for the management of symptomatic second-degree and thirddegree haemorrhoids (HubBLe): a multicentre, open-label, randomised controlled trial. Lancet 2016;388:356—64. http://dx.doi.org/10.1016/S0140-6736(16)30584-0
Background Optimum surgical intervention for low-grade haemorrhoids is unknown. Haemorrhoidal artery ligation (HAL) has been proposed as an efficacious, safe therapy while rubber band ligation (RBL) is a commonly used outpatient treatment. We compared recurrence after HAL versus RBL in patients with grade II-III haemorrhoids. Methods This multicentre, open-label, parallel group, randomised controlled trial included patients from 17 acute UK NHS trusts. We screened patients aged 18 years or older pre-
senting with grade II-III haemorrhoids. We excluded patients who had previously received any haemorrhoid surgery, more than one injection treatment for haemorrhoids, or more than one RBL procedure within 3 years before recruitment. Eligible patients were randomly assigned (in a 1:1 ratio) to either RBL or HAL with Doppler. Randomisation was computer-generated and stratified by centre with blocks of random sizes. Allocation concealment was achieved using a web-based system. The study was open-label with no masking of participants, clinicians, or research staff. The primary outcome was recurrence at 1 year, derived from the patient’s self-reported assessment in combination with resource use from their general practitioner and hospital records. Recurrence was analysed in patients who had undergone one of the interventions and been followed up for at least 1 year. This study is registered with the ISRCTN registry, ISRCTN41394716. Findings From Sept 9, 2012, to May 6, 2014, of 969 patients screened, 185 were randomly assigned to the HAL group and 187 to the RBL group. Of these participants, 337 had primary outcome data (176 in the RBL group and 161 in the HAL group). At 1 year post-procedure, 87 (49%) of 176 patients in the RBL group and 48 (30%) of 161 patients in the HAL group had haemorrhoid recurrence (adjusted odds ratio [aOR] 2.23, 95% CI 1.42—3.51; P = 0.0005). The main reason for this difference was the number of extra procedures required to achieve improvement (57 [32%] participants in the RBL group and 23 [14%] participants in the HAL group had a subsequent procedure for haemorrhoids). The mean pain 1 day after procedure was 3.4 (SD 2.8) in the RBL group and 4.6 (2.8) in the HAL group (difference −1.2, 95% CI −1.8 to −0.5; P = 0.0002); at day 7 the scores were 1.6 (2.3) in the RBL group and 3.1 (2.4) in the HAL group (difference −1.5, −2.0 to −1.0; P < 0.0001). Pain scores did not differ between groups at 21 days and 6 weeks. 15 individuals reported serious adverse events requiring hospital admission. One patient in the RBL group had a pre-existing rectal tumour. Of the remaining 14 serious adverse events, 12 (7%) were among participants treated with HAL and two (1%) were in those treated with RBL. Six patients had pain (one treated with RBL, five treated with HAL), three had bleeding not requiring transfusion (one treated with RBL, two treated with HAL), two in the HAL group had urinary retention, two in the HAL group had vasovagal upset, and one in the HAL group had possible sepsis (treated with antibiotics). Interpretation Although recurrence after HAL was lower than a single RBL, HAL was more painful than RBL. The difference in recurrence was due to the need for repeat bandings in the RBL group. Patients (and health commissioners) might prefer such a course of RBL to the more invasive HAL. Comments 1. Hemorrhoids are quite prevalent, implicating about one third of the overall population. While RBL is not expensive, can be performed in an ambulatory setting and has a low complication rate, the recurrence rate has been recognized to be relatively high [1], requiring further ligation sessions. Since HAL has been reported to have fewer recurrences, the authors sought to evaluate the best management plan in patients with grades II and III hemorrhoids in the first randomized trial comparing RBL to HAL. 2. If we consider RBL as a multiple stage procedure, there was no difference found in terms of efficacy between the
Press review two groups, recognizing that the level of pain was higher in HAL. 3. The overall analysis of results, in addition to cost-efficacy analysis, pleads in favor of RBL as long as the patient understands that management may require several sessions. Reference [1] Dis Colon Rectum 2004;47:1364—70. Thereaux J, Barsamian C, Bretault M, et al. pH monitoring of gastro-oesophageal reflux before and after laparoscopic sleeve gastrectomy. Br J Surg 2016;103:399—406. http://dx.doi.org/10.1002/bjs.10089
Background Gastro-oesophageal reflux disease (GORD) is a common obesity-related co-morbidity that is assessed objectively by 24-h pH monitoring. Some concerns have been raised regarding the risk of de novo GORD or exacerbation of preexisting GORD after laparoscopic sleeve gastrectomy. Here, 24-h pH monitoring was used to assess the influence of laparoscopic sleeve gastrectomy on post-operative GORD in obese patients with or without preoperativeents weof pati GORD. Methods From July 2012 to September 2014, all patients scheduled for laparoscopic sleeve gastrectomy were invited to participate in a prospective follow-up. Patients who underwent preoperative 24-h pH monitoring were asked to repeat the examination 6 months after operation. GORD was defined as an oesophageal pH < 4 for at least 4.2 per cent of the total time recorded. Results Of 89 patients, 76 had preoperative pH monitoring for GORD evaluation and 50 had postoperative reassessment. Patients without (group 1, 29 patients) or with (group 2, 21 patients) preoperative GORD were similar regarding age, sex ratio and body mass index. In group 1, the median (i.q.r.) total time at pH < 4 was significantly higher after surgery than before: 5.6 (2.5—9.5) versus 1.6 (0.7—2.9) per cent (P < 0.001). Twenty of the 29 patients experienced de novo GORD as determined by 24-h pH monitoring (P < 0.001). In group 2, total time at pH < 4 after surgery was no different from the preoperative value: 5.9 (3.9—10.7) versus 7.7 (5.2—10.3) per cent (P = 0.296). Conclusion Laparoscopic sleeve gastrectomy was associated with de novo GORD in over two-thirds of patients, but did not seem to exacerbate existing GORD. Comments 1. Whereas sleeve gastrectomy is currently recognized as valid and effective in the treatment of morbid obesity [1,2], and obesity is a recognized risk factor for gastroesophageal reflux disease (GERD) [3], the risk of GERD after sleeve gastrectomy remains poorly evaluated. The authors reported a statistically significant increased risk of de novo GERD, but without statistically significantly increasing the intensity of pre-existing GERD. 2. While the literature reports a risk of de novo GERD between 2 and 45% [4,5], the much higher rate reported in this study is related to the rigorous evaluation of the duration of pH < 4 on pH-metry, independently of whether GERD was symptomatic or not. 3. The absence of aggravation of pre-existing GERD is strengthened by other large sample studies showing
471 remission rates in 20% of patients and stabilization in 70% of patients [5], most likely related to decreased intraabdominal pressure as a consequence of weight-loss. 4. Among the possible explanations for de novo GERD after sleeve gastrectomy are the anatomical modification of the angle of His, transection of muscular fibers, increased intra-gastric pressure, decreased lower esophageal sphincter pressure, and the onset of hiatal hernia. 5. The consequences of de novo GERD on the long-term risk of esophageal cancer are unknown. The pre-operative presence of GERD and a fortiori the presence of intestinal metaplasia should plead in favor of gastric bypass. References [1] Obes Surg 2013;23:2013—17. [2] Ann Surg 2013;258:690—4. [3] Surg Obes Relat Dis 2013;9:159—91. [4] JAMA Surg 2014;149:328—34. [5] Surg Obes Relat Dis 2015;11:222—8. Hackert T, Sachsenmaier M, Hinz U, et al. Locally advanced pancreatic cancer: neoadjuvant therapy with folfirinox results in resectability in 60% of the patients. Ann Surg 2016;264:457—63. http://dx.doi.org/10.1097/SLA. 0000000000001850
Objective For patients with locally advanced and unresectable pancreatic cancer (PDAC), neodadjuvant treatment and consecutive surgical exploration have been studied during the last decade with various neoadjuvant therapies including chemotherapy and combinations with radiation. Aim of the study was the evaluation of neoadjuvant therapy with a focus on Folfirinox. Methods All consecutive patients undergoing surgery for PDAC after neoadjuvant treatment were analyzed (clinicopathological characteristics, secondary resection rates, outcome). Patients receiving Folfirinox were compared with other treatment regimens. Results Between December 2001 and June 2015, 575 patients received neoadjuvant treatment and were scheduled for resection after re-staging. A successful resection was achieved in 292 patients (50.8%). Resection rates following Folfirinox were 61% (76/125 patients) compared with 46% (150/322 patients) after gemcitabine and radiation, and 52% (66/128 patients) after other treatments (P = 0.026). Median overall survival was 15.3 months after resection vs 8.5 months after exploration alone (P < 0.0001). Subgroup median survival was 16.0 months (Folfirinox) vs 16.5 months (gemcitabine) and 14.5 months (others) with 3-year survival of 28.1%, 23.2%, and 19.7%, respectively (P = 0.8582). By multivariable analysis, Folfirinox was confirmed to be independently associated with a favorable prognosis. Conclusions Folfirinox is a valuable treatment option in the neoadjuvant therapy of PDAC. From the present data, which represent the largest available study population to date, Folfirinox seems to be the most effective protocol resulting in a significantly better secondary resection rate and overall survival than other treatments. It should be considered in all patients fit for this regimen and consecutive surgical exploration.
472 Comments 1. In spite of improvement in medical therapy, surgery remains the only treatment with curative intent for pancreatic cancer. Because of vascular or adjacent organ involvement, surgical possibilities are limited for PDAC. The authors suggest that neo-adjuvant chemotherapy, and in particular, Folfirinox, can obtain a 61% resectability rate and increase overall survival. 2. As of today, there are no randomized studies available concerning the benefits of Folfirinox in PDAC, but several trials are underway, notably one in France. One recent meta-analysis found that secondary resection was possible in only 43% of patients after Folfirinox. 3. Of note, even if these authors highlight a high surgical resection rate, the R0 resection rate was only 40.8% after Folfirinox (vs. 31.3% after gemcitabine and 27.3% after other treatment options, P = 0.041), this low rate being responsible for the moderate benefit in overall survival.
C. Mariette, S. Benoist 4. The two groups re not comparable as concerned the study periods under analysis as well as age, both being substantial causes of bias. Cancer-specific survival analysis would have been more pertinent. 5. This study has several potential biases that could have influenced outcome, notably (i) absence of precisions concerning the number of patients included in the treatment plan who, after that period, were still not amenable to surgery, (ii) a highly probable mix of borderline and locally advanced tumors, (iii) a selection bias, (iv) a shorter follow-up in the Folfirinox group (only recently available in the therapeutic armamentarium, lessening the chances that recurrence or death could occur), (v) survival was calculated starting with the date of surgery (starting with the beginning of treatment would have been more correct), (vi) 51 patients with initial metastatic disease were included. Déclaration de liens d’intérêts Les auteurs déclarent ne pas avoir de liens d’intérêts.
Available online at
ScienceDirect www.sciencedirect.com
Press review C. Mariette a,∗, S. Benoist b a
Service de chirurgie digestive et générale, hôpital C.-Huriez, CHRU, place de Verdun, 59037 Lille cedex, France b Service de chirurgie digestive, hôpital du Kremlin-Bicêtre, 78, rue du Général-Leclerc, 94275 Le Kremlin-Bicêtre, France Available online 16 November 2016 Cousin F, Ortega-Deballon P, Bourredjem A, et al. Diagnostic accuracy of procalcitonin and C-reactive protein for the early diagnosis of intra-abdominal infection after elective colorectal surgery: a meta-analysis. Ann Surg 2016;264:252—6. http://dx.doi.org/10.1097/SLA. 0000000000001545
Objective Intra-abdominal infections (IAIs) after elective colorectal surgery impact significantly the short- and long-term outcomes. In the era of fast-track surgery, they often come to light after discharge from hospital. Early diagnosis is therefore essential. C-reactive protein levels have proved to be accurate in this setting. Procalcitonin has been evaluated in several studies with conflicting results. This meta-analysis aimed to compare the predictive abilities of C-reactive protein and procalcitonin in the occurrence of IAIs after elective colorectal surgery. Methods This meta-analysis included studies analyzing C-reactive protein and/or procalcitonin levels at postoperative days 2, 3, 4, and/or 5 as markers of intra-abdominal infection after elective colorectal surgery. Methodological quality was assessed by the QUADAS2 tool. The area under the curve summary receiver-operating characteristic was calculated for each day and each biomarker, using a random-effects model in cases of heterogeneity. Results The meta-analysis included 11 studies (2692 patients). An IAI occurred in 8.9% of the patients. On postoperative day 3, area under the curve was 0.80 (95% CI, 0.76—0.85) for C-reactive protein and 0.78 (95% CI, 0.68—0.87) for procalcitonin. On postoperative day 5, their predictive accuracies were 0.87 (95% CI, 0.80—0.93) and 0.90 (95% CI, 0.82—0.98),
∗
Corresponding author. E-mail address: [email protected] (C. Mariette).
1878-7886/$ — see front matter http://dx.doi.org/10.1016/j.jviscsurg.2016.09.018
respectively. The accuracy of C-reactive protein and procalcitonin did not differ at any postoperative day. Conclusions Levels of inflammatory markers under the cutoff value between postoperative days 3 and 5 ensure safe early discharge after elective colorectal surgery. Procalcitonin seems not to have added value as compared to C-reactive protein in this setting. Comments 1. This meta-analysis confirms the results of the French prospective study organized by the same authors that showed that procalcitonin (PCT) was not useful for the early detection of infectious complications after colorectal surgery [1]. 2. This meta-analysis, however, has several methodological biases. Firstly, the heterogeneity in the definition of the main end-point, i.e. IAI. Some of the studies only included anastomotic leaks, while others also included simple intra-abdominal collections seen on imaging. Moreover, in all the studies, the manner to search for IAI was not the same, and therefore, some might have escaped diagnosis. Last, in all the studies except one, the investigators whose role was to look for IAI postoperatively were aware of the values of CRP and PCT, which might have encouraged the investigators to look for IAI in asymptomatic patients with high values, thus increasing the positive and negative predictive values of these two investigations. 3. The main question for which this meta-analysis does not provide any answer is to know whether early detection of IAI in pauci-symptomatic patients can change ulterior management. Reference [1] Ann Surg 2016;263:967—72. Yamada T, Okabayashi K, Hasegawa H, et al. Meta-analysis of the risk of small bowel obstruction following open or laparoscopic colorectal surgery. Br J Surg 2016;103:493—503. http://dx.doi.org/10.1002/bjs.10105
466
C. Mariette, S. Benoist
Background One of the potential advantages of laparoscopic compared with open colorectal surgery is a reduction in postoperative bowel obstruction events. Early reports support this proposal, but accumulated evidence is lacking. Methods A systematic review and meta-analysis was performed of randomized clinical trials and observational studies by searching the PubMed and Cochrane Library databases from 1990 to August 2015. The primary outcomes were early and late postoperative bowel obstruction following laparoscopic and open colorectal surgery. Both ileus and bowel obstruction were defined as a postoperative bowel obstruction. Subgroup and sensitivity analyses were performed, and a random-effects model was used to account for the heterogeneity among the studies. Results Twenty-four randomized clinical trials and 88 observational studies were included in the meta-analysis; 106 studies reported early outcome and 12 late outcome. Collectively, these studies reported on the outcomes of 148—392 patients, of whom 58—133 had laparoscopic surgery and 90—259 open surgery. Compared with open surgery, laparoscopic surgery was associated with reduced rates of early (odds ratio 0.62, 95 per cent c.i. 0.54 to 0.72; P < 0.001) and late (odds ratio 0.61, 0.41 to 0.92; P = 0.019) postoperative bowel obstruction. Weighted mean values for early postoperative bowel obstruction were 8 (95 per cent c.i. 6 to 10) and 5 (3 to 7) per cent for open and laparoscopic surgery respectively, and for late bowel obstruction were 4 (2 to 6) and 3 (1 to 5) per cent respectively. Conclusion The reduction in postoperative bowel obstruction demonstrates an advantage of laparoscopic surgery in patients with colorectal disease. Comments 1. In patients with early bowel obstruction, the authors did not distinguish between post-operative ileus and true obstruction due to adhesions or bands. Moreover, there were no precisions given concerning the diagnosis of obstruction: was it based on clinical, surgical or radiological findings? 2. With respect to late bowel obstruction, there was no information concerning the mean follow-up in the different studies. It is likely that this period was short, as the 4% post-laparotomy rate appears to be extremely low. A longer follow-up period would have permitted inclusion of more events, and therefore a more precise evaluation of the magnitude of the effect. 3. While this meta-analysis is not perfect methodologically speaking, it constitutes another argument in favor of the laparoscopic approach in colorectal surgery.
Bains SJ, Mahic M, Myklebust TÅ, et al. Aspirin as secondary prevention in patients with colorectal cancer: an unselected population-based study. J Clin Oncol 2016;34:2501—2508. http://dx.doi.org/10.1200/JCO.2015. 65.3519
Purpose Regular use of aspirin (acetylsalicylic acid) is associated with reduced incidence and mortality of colorectal cancer (CRC). However, aspirin as primary prevention is debated because of the risk of hemorrhagic adverse effects. Aspirin as secondary prevention may be more justified from
a risk-benefit perspective. We have examined the association between aspirin use after the diagnosis of CRC with CRC-specific survival (CSS) and overall survival (OS). Materials and methods An observational, population-based, retrospective cohort study was conducted by linking patients diagnosed with CRC from 2004 through 2011 (Cancer Registry of Norway) with data on their aspirin use (The Norwegian Prescription Database). These registries cover more than 99% of the Norwegian population and include all patients in an unselected and consecutive manner. Exposure to aspirin was defined as receipt of aspirin prescriptions for more than 6 months after the diagnosis of CRC. Multivariable Cox-proportional hazard analyses were used to model survival. The main outcome measures of the study were CSS and OS. Results A total of 23,162 patients diagnosed with CRC were included, 6102 of whom were exposed to aspirin after the diagnosis of CRC (26.3%). The median follow-up time was 3.0 years. A total of 2071 deaths (32.9%, all causes) occurred among aspirin-exposed patients, of which 1158 (19.0%) were CRC specific. Among unexposed patients (n = 17,060), there were 7218 deaths (42.3%), of which 5375 (31.5%) were CRC specific. In multivariable analysis, aspirin exposure after the diagnosis of CRC was independently associated with improved CSS (hazard ratio [HR], 0.85; 95% CI, 0.79 to 0.92) and OS (HR, 0.95; 95% CI, 0.90 to 1.01). Conclusion Aspirin use after the diagnosis of CRC is independently associated with improved CSS and OS. Comments 1. The interpretation of the conclusions warrants caution because of the many methodological biases in this study. Firstly, the patients in each group were not comparable, particularly for the major prognostic factors such as disease stage or tumor differentiation. Effectively, tumors in the aspirin group were less locally advanced and more differentiated than in the control group. Moreover, in addition to surgery, there was no information concerning the treatment for cancer and in particular, concerning chemotherapy. Lastly, several prognostic factors such as CEA, the R0 resection rate, administration of chemoradiation in patients with rectal cancer, the site of synchronous metastases if present, were not included in the multivariable analysis. 2. From the methodological point of view, since the two groups were not absolutely comparable, the authors should have performed a propensity score using logistic regression including the variables that differed between the two groups, and then either include this score in the multivariable model, or match the patients using the score. 3. It is difficult to determine whether aspirin use after the diagnosis of CRC improved cancer-specific survival, as it was in patients who took aspirin before and after diagnosis that the greater benefit was observed. It is also probable that there was a greater benefit for patients who took aspirin before diagnosis (vs. after) since cancerspecific survival was not improved in the subgroup of patients who only took aspirin after diagnosis was made.
Fazi M, Giudici F, Luceri C, et al. Long-term results and recurrence-related risk factors for Crohn disease in patients undergoing side-to-side isoperistaltic strictureplasty. JAMA
Press review Surg 2016;151:452—60. jamasurg.2015.4552
467 http://dx.doi.org/10.1001/
Importance Side-to-side isoperistaltic strictureplasty (SSIS) is useful in patients undergoing surgery for Crohn disease (CD) to avoid wide small-bowel resections. To our knowledge, there are no definitive data regarding its recurrence risk factors. Objective To evaluate the results obtained in a monocentric population of patients with CD who have undergone SSIS. Design, setting, and participants From August 1996 to March 2010, 91 patients with CD underwent SSIS in our center. In this prospective observational study, side-to-side isoperistaltic strictureplasty was according the Michelassi technique in 69 patients and the Tonelli technique in 22 patients. Factors relating to the patient and the CD, surgery, and pharmacological therapy during the preoperative and perioperative periods were evaluated in association with medical or surgical recurrence. Exposure Side-to-side isoperistaltic strictureplasty. Main outcomes and measures The recurrence-free curve was estimated using KaplanMeier analysis. Patients were stratified into cohorts in relation to the considered categorical variables and data were compared by using the Mantel-Cox log-rank test. Cox proportional hazard regression analysis was used to set up a predictive model simultaneously exploring the effects of all independent variables on a dichotomous outcome recurrence in relation to time. Results Among the 91 patients, the mean (SD) age was 39.5 (11.2) years and preoperative disease duration was 97.9 (85.8) months; 83 patients (91.2%) were followed up, of whom 37 (44.58%) experienced a recurrence at a mean (SD) of 55.46 (36.79) months after surgery (range, 9—140 months). The recurrence in the SSIS site at a mean (SD) of 48.25 (29.94) months after surgery affected 24 of 83 patients (28.9%), 9 being medical and 15 being surgical recurrence. Recurrence in the SSIS was statistically significantly associated with the time elapsed between diagnosis and surgery (P = .03). A borderline association between family history of CD and surgical recurrence (P = .054) was also found. Multivariate analysis identified the age at diagnosis (2 = 5.56; P = .02) and at surgery (2 = 7.77; P = .005), family history (2 = 6.26; P = .01), and smoking habit (2 = 10.06; P = .007) as independent risk factors for recurrence. Conclusions and relevance In the short-term, SSIS leads to a resolution of symptoms in more than 90% of cases and the recurrence rate in the SSIS area is acceptable, even after long-term follow-up. Comments 1. This study confirms that there are very few indications for strictureplasty; in this specialized center for CD, only 91 operations with strictureplasty were performed over 12 years, i.e. fewer than eight per year. It would therefore have been interesting to know the total amount of patients undergoing operation for CD during the same period. 2. All the patients included in this series had severe disease with multiple site involvement and more than half had already undergone surgery for their CD, a well-known risk factor for recurrence after surgery [1]. Only 1/3 of these patients had received immuno-suppressive or antiTNF treatment before surgery. This means some of the
patients in this series received sub-optimal treatment since these patients would not have undergone operation today or, in any case, only after failure of several lines of therapy. Likewise, giving only Pentasa® for patients at high risk of recurrence does not seem to be appropriate, since today, most of these patients would have immuno-suppressive therapy [2]. 3. Finally, the clinical impact of this narrative study is practically nil; the results of this study will probably not lead to increased indications for long side-to-side strictureplasties; they only confirm those of a multicenter study published in 2007 [3]. Moreover, the 45% recurrence rate is comparable to that reported after resection [4]. Lastly, the risk factors found here are already well-known risk factors for recurrence. With the emergence of new treatments, the risk of short bowel syndrome is rare and the situations where strictureplasty should be preferred over resection are not common. References [1] Dis Colon Rectum 2009;52:919—27. [2] Am J Gastroenterol 2010;105:1158—64. [3] Dis Colon Rectum 2007;50:277—84. [4] Inflam Bowel Dis 2013;19:7—14. Sadot E, Lee SY, Sofocleous CT, et al. Hepatic resection or ablation for isolated breast cancer liver metastasis: a case-control study with comparison to medically treated patients. Ann Surg 2016;264:147—54. http://dx.doi.org/ 10.1097/SLA.0000000000001371
Objective To evaluate the efficacy of surgical treatment for patients with isolated breast cancer liver metastases (BCLM). Background Single-arm retrospective studies have shown promising results associated with surgery for isolated BCLM; however, this treatment remains controversial and its role is not welldefined. Methods A review of 2150 patients with BCLM who underwent treatment in a single institution was conducted, and 167 (8%) patients with isolated BCLM were identified. A case-control study was conducted to compare outcomes in patients with isolated BCLM who underwent surgery and/or ablation to patients who underwent conventional medical therapy. Results A total of 167 patients were included (surgery/ablation: 69; medical: 98), with a median follow-up for survivors of 73 months. Patients in the surgical cohort more frequently had estrogen receptor-positive tumors and received adjuvant chemotherapy and radiotherapy for their primary breast tumor. The hepatic tumor burden was less and the interval from breast cancer diagnosis to BCLM was significantly longer (53 vs 30 months) in the surgical cohort. Patients undergoing surgical treatment had a median recurrence-free interval of 28.5 months (95% confidence interval (CI): 19—38) with 10 patients (15%) recurrence free after 5 years. There was no significant difference in overall survival (OS) between the surgical and medical cohorts (median OS: 50 vs 45 months; 5-year OS: 38% vs 39%). Conclusions Hepatic resection and/or ablation was not associated with a survival advantage. However, significant recurrencefree intervals can be accomplished with surgical treatment. Surgical intervention might be considered in highly selected
468
C. Mariette, S. Benoist
patients with the goal of providing time off of systemic chemotherapy. Comments 1. This study rekindles the debate on the management of patients with BCLM and is in contradiction with the results of initial studies that were strongly in favor of surgery. Notwithstanding, the outcome of surgery here is rather good; the 40% 5-year survival rate is comparable to that reported in other surgical series [1—3]. 2. However, one could be led to think that surgery is not as good as medical treatment since survival was not improved in the surgical group even though patients in this group had less aggressive disease. A recent metaanalysis including 1341 patients with metastatic breast cancer showed that prognosis improved over the 20-year period of inclusion, a period during which surgery was performed less and less often [4]. This suggests that improved prognosis was due to new medications rather than surgery. 3. In previous studies [1—3], it was shown that the best indications for surgery were good control of BCLM by induction chemotherapy. In this study, however, patients undergoing surgery for their BCLM apparently did not receive pre-operative chemotherapy, which could have worsened the results for surgery. 4. Regrettably, the authors did not evaluate quality of life. Since patients in the surgery group had an interval median follow-up period of 25 months without chemotherapy (compared to patients in the chemotherapy group who were treated sans interruption), which could have greatly impacted quality of life. Therefore, surgery might still be the best option for BCLM, since it is associated with better quality of life without decreasing survival. References [1] Ann Surg Oncol 2004;11:869—74. [2] Ann Surg 2006;244:897—907. [3] Eur J Cancer 2011;47:2282—90. [4] Breast Cancer Res Treat 2010;119:621—31. Hof J, Wertenbroek MW, Peeters PM, et al. Prospective outcomes after resection and/or radiofrequency ablation for recurrence after treatment of colorectal liver metastases. Br J Surg 2016;103:1055—62. http://dx.doi.org/10.1002/ bjs.10162
Background Repeat liver resection for colorectal liver metastases (CRLMs) is possible in a limited number of patients, with radiofrequency ablation (RFA) as an alternative for unresectable CRLMs. The aim of this study was to analyse survival rates with these interventions. Methods This was a database analysis of patients who underwent first and repeat interventions for synchronous and metachronous CRLMs between 2000 and 2013. Descriptive and survival statistics were calculated. Results Among 431 patients who underwent resection or RFA for CRLMs, 305 patients developed recurrences for which 160 repeat interventions (resection and/or RFA or ablative radiotherapy) were performed. In total, after 707 first or repeat interventions, 516 recurrences (73.0 per cent) developed, of which 276 were retreated curatively. At the time of first intervention, independent risk factors for death were
lymph node-positive primary tumour (hazard ratio [HR] 1.40; P = 0.030), more than one CRLM (HR 1.53; P = 0.007), carcinoembryonic antigen level exceeding 200 ng/ml (HR 1.89; P = 0.020) and size of largest CRLM greater than 5 cm (HR 1.54; P = 0.014). The 5-year overall survival rates for liver resection and percutaneous RFA as first intervention were 51.9 and 53 per cent, with a median overall survival of 65.0 (95 per cent c.i. 47.3 to 82.6) and 62.1 (52.2 to 72.1) months, respectively. Conclusion RFA had good oncological outcomes in patients with unresectable CRLMs. Radiofrequency ablation is progressively more applied with each additional intervention. Comments 1. The only strong point in this series is that analysis was conducted with intention to treat. Of particular interest, iterative intervention was possible in only 53% of patients with recurrent CRLM in this unselected population. 2. It is difficult to conclude that iterative surgery increased overall survival since there was no control group treated with chemotherapy alone. Moreover, oncological results in this series were good but not exceptional, since a 48% 5-year survival rate is comparable to those reported in other series where iterative interventions were less frequent [1,2]. 3. The main message of this study is to show that radiofrequency ablation allows iterative interventions. Effectively, radiofrequency ablation alone was used firstline in 30% of patients and in 50% beginning with the second re-intervention. References [1] Lancet Oncol 2013;14:1208—15. [2] J Clin Oncol 2007;25:4575—80. Fang X, Baillargeon JG, Jupiter DC. Continued antiplatelet therapy and risk of bleeding in gastrointestinal procedures: a systematic review. J Am Coll Surg 2016;222:890—905. http://dx.doi.org/10.1016/j.jamcollsurg.2016.01.053
Background Management of peri-operative antiplatelet medications in gastrointestinal (GI) surgery is challenging. The risk of intraoperative and postoperative bleeding is associated with perioperative use of antiplatelet medication. However, cessation of these drugs may be unsafe for patients who are required to maintain antiplatelet use due to cardiovascular conditions. The objective of this systematic review was to compare the risk of intraoperative or postoperative bleeding among patients who had GI surgery while on continuous antiplatelet therapy (aspirin, clopidogrel, or dual therapy) with the risk among those not taking continuous antiplatelet medication. Study design We reviewed articles published between January 2000 and July 2015 from the Medline Ovid and Cumulative Index to Nursing and Allied Health Literature (CINAHL) databases. Studies involving any GI procedures were included if the articles met our inclusion criteria (listed in Methods). The following keywords were used for the search: clopidogrel, Plavix, aspirin, antiplatelet, bleeding, hemorrhage, and digestive system surgical procedures. Quality of the studies was assessed, depending on their study design, using the Newcastle-Ottawa score or the Cochrane Collaboration’s tool for assessing risk of bias.
Press review Results Twenty-two studies were eligible for inclusion in the systematic review. Five showed that the risk of intraoperative bleeding or postoperative bleeding among patients who had GI surgery while on continuous antiplatelet therapy was higher compared that for those not on continuous therapy. The remaining 17 studies reported that there was no statistically significant difference in the risks of bleeding between the continuous antiplatelet therapy group and the group without continuous antiplatelet therapy. Conclusions The risk of bleeding associated with GI procedures in patients receiving antiplatelet therapy was not significantly higher than in patients with no antiplatelet or interrupted antiplatelet therapy. Comments 1. The highlight of this study is that it goes against the dogma recommending that antiplatelet therapy (APT), and particularly, aspirin, should be discontinued before surgery or gastro-intestinal endoscopy, an idea that has prevailed for decades. The risk-benefit ratio favors continuation of APT before, during and after these procedures. 2. Regrettably, no meta-analysis was performed; a statistical compilation of events would have provided a more precise idea of the risk, evaluated by an odds ratio, whether for hemorrhage in the group with APT, or ischemia in the group with discontinuation of APT. 3. It might have been interesting to have a precise evaluation of the cardiovascular risk in case of discontinued APT. 4. The level of evidence is high for aspirin but much lower for clopidrogel as there were only four studies dedicated to this APT. Consequently, while waiting for further studies, it seems reasonable to follow the recommendations laid down by the American Society for Gastrointestinal Endoscopy (ASGE) [1]: continue aspirin, but discontinue clopidrogel 7 to 10 days before gastro-intestinal surgery in patients with a low thrombo-embolic risk and weigh the risk-benefit ratio for patients at high risk. 5. The limitations of this study include the lack of information about: the doses used, the period of exposure before surgery, identification of subgroups for which this strategy should be modeled as well as quantitative analysis of results. Reference [1] Gastrointest Endosc 2009;70:1060—70. Barazanchi AW, Fagan PV, Smith BB, Hill AG. Routine neurectomy of inguinal nerves during open onlay mesh hernia repair: a meta-analysis of randomized trials. Ann Surg 2016;264:64—72. http://dx.doi.org/10.1097/SLA. 0000000000001613
Objective The aim of the study was to establish whether an inguinal neurectomy at the time of hernia repair would reduce the risk of postoperative pain for open tension-free sutured mesh repair. Background Inguinal hernia repair is a common operative procedure. The development of postoperative pain is uncommon, but at times debilitating. The role of inguinal neurectomy is currently unknown, with no single large study available, and previous reviews included only a few heterogeneous studies.
469 Methods Relevant randomized trials were identified from searches of MEDLINE, EMBASE, and EBM Review databases until October 2014. Meta-analysis was performed based on Cochrane Methods using RevMan v5.3 software. Pain, pain scores, sensory changes, and complications over short (half to < 3 months), mid (3 to < 12 months), and long term (≥ 12 months) were recorded. Results All included studies performed Lichtenstein hernia repair. Eleven studies on 1031 patients showed significant reduction in pain with neurectomy for short (RR = 0.61, 0.40-0.93) and midterm (RR = 0.30, 0.20—0.46), but not for long term (RR = 0.50, 0.25—1.01). Three studies (270 patients) showed significantly reduced short-term pain (RR = 0.69, 0.52—0.90). No studies included genitofemoral neurectomy. Rates of hematoma, infection, urinary retention, and recurrence were not different between groups. Conclusions Routine ilioinguinal neurectomy during Lichtenstein-type herniorrhaphy seems to be a safe and effective method to reduce pain in the short and midterm, but may have little long-term impact. Iliohypogastric neurectomy seems to reduce pain in at least the short term. Comments 1. The Lichtenstein open repair is the standard inguinal hernia repair technique for the American College of Surgeons [1], having been shown to be the safest of inguinal hernia repair techniques [2]. The European Hernia Society proposes laparoscopic repair but the learning curve is long [3]. 2. While recurrence is low, the principal complication reported is chronic post-operative pain [4], iatrogenic nerve injury being the most frequently reported promoting factor [5]. 3. There is no explanation for increased pain when nerves are spared. However, inflammation, entrapment or perineural fibrosis have been incriminated. The role of prosthetic material is not well known. References [1] Hernia 2004;8:1—7. [2] Ann Surg 2012;255:784—88. [3] Hernia 2014;18:151—63. [4] Ann Surg 2006;244:212—9. [5] Lancet 2006;367:1618—25. Kadry Z, Schaefer EW, Shah RA, et al. Portal hypertension: an underestimated entity? Ann Surg 2016;263:986—91. http://dx.doi.org/10.1097/SLA.0000000000001299
Objective The aim of this study is to evaluate portal hypertension as an independent risk factor in general surgical procedures. Background Data on the impact of portal hypertension in general surgical outcomes has been limited. Published literature has focused mainly on its effect in liver surgery. The Child Pugh score and Model for End Stage Liver Disease are utilized for surgical risk assessment in liver disease but they do not accurately reflect degree of portal hypertension. Methods From 2005 to 2012, patients with esophageal varices (EV) in the National Surgical Quality Improvement Program (NSQIP) formed the portal hypertension cohort, and were case matched to patients without esophageal varices (NEV)
470
C. Mariette, S. Benoist
based on sex, age, surgery type, and year of operation. Thirty day mortality and morbidity were analyzed using generalized estimating equations for binary outcomes. EV patients were also dichotomized by Model for End Stage Liver Disease (MELD) score (≤15 vs >15) and compared with NEV patients. Results One thousand five hundred and seventy-four EV patients were matched to 3148 NEV patients. In multivariable analysis, EV patients had a 3.01 higher odds of 30 day mortality (P < 0.001) and 1.28 higher odds of complications (P < 0.001) compared with NEV patients. EV patients with MELD >15 had 4.64 higher odds of death within 30 days (P < 0.001) and had 1.75 higher odds of complications within 30 days (P < 0.001) compared with NEV patients; EV patients with MELD 15 or less had 1.95 higher odds of 30 day mortality (P < 0.001) compared with NEV patients. Conclusions Portal hypertension is associated with a significant mortality and morbidity risk in general surgery, and should not be underestimated even in patients with MELD 15 or less where the early mortality risk remained significant. Comments 1. Portal hypertension (PHT) is associated with increased mortality and morbidity in liver surgery, particularly in patients with cirrhosis [1,2]. Here the authors underscored that PHT significantly increased post-operative mortality and morbidity after general, non-hepatic surgery. 2. The value of Child-Pugh’s score is limited by the subjectivity in the evaluation of certain variables entering the score such as the degree of ascites and encephalopathy. The MELD score, conversely, does not explore any of the components of PHT and is limited to three variables: total bilirubinemia, international normalized ratio (INR) and creatininemia, which can be affected by non-liver causes. The results of this study were that the MELD score influences the risk of post-operative death in patients with EV, even in patients where the MELD score ≤15. 3. Regrettably, PHT was evaluated only by the presence of EV; some patients could have had PHT without EV and therefore might have been included in the NEV group. References [1] Gastroenterology 1996;111:1018—22. [2] Ann Surg 2006;243:373—9. Brown SR, Tiernan JP, Watson AJ, et al. Haemorrhoidal artery ligation versus rubber band ligation for the management of symptomatic second-degree and thirddegree haemorrhoids (HubBLe): a multicentre, open-label, randomised controlled trial. Lancet 2016;388:356—64. http://dx.doi.org/10.1016/S0140-6736(16)30584-0
Background Optimum surgical intervention for low-grade haemorrhoids is unknown. Haemorrhoidal artery ligation (HAL) has been proposed as an efficacious, safe therapy while rubber band ligation (RBL) is a commonly used outpatient treatment. We compared recurrence after HAL versus RBL in patients with grade II-III haemorrhoids. Methods This multicentre, open-label, parallel group, randomised controlled trial included patients from 17 acute UK NHS trusts. We screened patients aged 18 years or older pre-
senting with grade II-III haemorrhoids. We excluded patients who had previously received any haemorrhoid surgery, more than one injection treatment for haemorrhoids, or more than one RBL procedure within 3 years before recruitment. Eligible patients were randomly assigned (in a 1:1 ratio) to either RBL or HAL with Doppler. Randomisation was computer-generated and stratified by centre with blocks of random sizes. Allocation concealment was achieved using a web-based system. The study was open-label with no masking of participants, clinicians, or research staff. The primary outcome was recurrence at 1 year, derived from the patient’s self-reported assessment in combination with resource use from their general practitioner and hospital records. Recurrence was analysed in patients who had undergone one of the interventions and been followed up for at least 1 year. This study is registered with the ISRCTN registry, ISRCTN41394716. Findings From Sept 9, 2012, to May 6, 2014, of 969 patients screened, 185 were randomly assigned to the HAL group and 187 to the RBL group. Of these participants, 337 had primary outcome data (176 in the RBL group and 161 in the HAL group). At 1 year post-procedure, 87 (49%) of 176 patients in the RBL group and 48 (30%) of 161 patients in the HAL group had haemorrhoid recurrence (adjusted odds ratio [aOR] 2.23, 95% CI 1.42—3.51; P = 0.0005). The main reason for this difference was the number of extra procedures required to achieve improvement (57 [32%] participants in the RBL group and 23 [14%] participants in the HAL group had a subsequent procedure for haemorrhoids). The mean pain 1 day after procedure was 3.4 (SD 2.8) in the RBL group and 4.6 (2.8) in the HAL group (difference −1.2, 95% CI −1.8 to −0.5; P = 0.0002); at day 7 the scores were 1.6 (2.3) in the RBL group and 3.1 (2.4) in the HAL group (difference −1.5, −2.0 to −1.0; P < 0.0001). Pain scores did not differ between groups at 21 days and 6 weeks. 15 individuals reported serious adverse events requiring hospital admission. One patient in the RBL group had a pre-existing rectal tumour. Of the remaining 14 serious adverse events, 12 (7%) were among participants treated with HAL and two (1%) were in those treated with RBL. Six patients had pain (one treated with RBL, five treated with HAL), three had bleeding not requiring transfusion (one treated with RBL, two treated with HAL), two in the HAL group had urinary retention, two in the HAL group had vasovagal upset, and one in the HAL group had possible sepsis (treated with antibiotics). Interpretation Although recurrence after HAL was lower than a single RBL, HAL was more painful than RBL. The difference in recurrence was due to the need for repeat bandings in the RBL group. Patients (and health commissioners) might prefer such a course of RBL to the more invasive HAL. Comments 1. Hemorrhoids are quite prevalent, implicating about one third of the overall population. While RBL is not expensive, can be performed in an ambulatory setting and has a low complication rate, the recurrence rate has been recognized to be relatively high [1], requiring further ligation sessions. Since HAL has been reported to have fewer recurrences, the authors sought to evaluate the best management plan in patients with grades II and III hemorrhoids in the first randomized trial comparing RBL to HAL. 2. If we consider RBL as a multiple stage procedure, there was no difference found in terms of efficacy between the
Press review two groups, recognizing that the level of pain was higher in HAL. 3. The overall analysis of results, in addition to cost-efficacy analysis, pleads in favor of RBL as long as the patient understands that management may require several sessions. Reference [1] Dis Colon Rectum 2004;47:1364—70. Thereaux J, Barsamian C, Bretault M, et al. pH monitoring of gastro-oesophageal reflux before and after laparoscopic sleeve gastrectomy. Br J Surg 2016;103:399—406. http://dx.doi.org/10.1002/bjs.10089
Background Gastro-oesophageal reflux disease (GORD) is a common obesity-related co-morbidity that is assessed objectively by 24-h pH monitoring. Some concerns have been raised regarding the risk of de novo GORD or exacerbation of preexisting GORD after laparoscopic sleeve gastrectomy. Here, 24-h pH monitoring was used to assess the influence of laparoscopic sleeve gastrectomy on post-operative GORD in obese patients with or without preoperativeents weof pati GORD. Methods From July 2012 to September 2014, all patients scheduled for laparoscopic sleeve gastrectomy were invited to participate in a prospective follow-up. Patients who underwent preoperative 24-h pH monitoring were asked to repeat the examination 6 months after operation. GORD was defined as an oesophageal pH < 4 for at least 4.2 per cent of the total time recorded. Results Of 89 patients, 76 had preoperative pH monitoring for GORD evaluation and 50 had postoperative reassessment. Patients without (group 1, 29 patients) or with (group 2, 21 patients) preoperative GORD were similar regarding age, sex ratio and body mass index. In group 1, the median (i.q.r.) total time at pH < 4 was significantly higher after surgery than before: 5.6 (2.5—9.5) versus 1.6 (0.7—2.9) per cent (P < 0.001). Twenty of the 29 patients experienced de novo GORD as determined by 24-h pH monitoring (P < 0.001). In group 2, total time at pH < 4 after surgery was no different from the preoperative value: 5.9 (3.9—10.7) versus 7.7 (5.2—10.3) per cent (P = 0.296). Conclusion Laparoscopic sleeve gastrectomy was associated with de novo GORD in over two-thirds of patients, but did not seem to exacerbate existing GORD. Comments 1. Whereas sleeve gastrectomy is currently recognized as valid and effective in the treatment of morbid obesity [1,2], and obesity is a recognized risk factor for gastroesophageal reflux disease (GERD) [3], the risk of GERD after sleeve gastrectomy remains poorly evaluated. The authors reported a statistically significant increased risk of de novo GERD, but without statistically significantly increasing the intensity of pre-existing GERD. 2. While the literature reports a risk of de novo GERD between 2 and 45% [4,5], the much higher rate reported in this study is related to the rigorous evaluation of the duration of pH < 4 on pH-metry, independently of whether GERD was symptomatic or not. 3. The absence of aggravation of pre-existing GERD is strengthened by other large sample studies showing
471 remission rates in 20% of patients and stabilization in 70% of patients [5], most likely related to decreased intraabdominal pressure as a consequence of weight-loss. 4. Among the possible explanations for de novo GERD after sleeve gastrectomy are the anatomical modification of the angle of His, transection of muscular fibers, increased intra-gastric pressure, decreased lower esophageal sphincter pressure, and the onset of hiatal hernia. 5. The consequences of de novo GERD on the long-term risk of esophageal cancer are unknown. The pre-operative presence of GERD and a fortiori the presence of intestinal metaplasia should plead in favor of gastric bypass. References [1] Obes Surg 2013;23:2013—17. [2] Ann Surg 2013;258:690—4. [3] Surg Obes Relat Dis 2013;9:159—91. [4] JAMA Surg 2014;149:328—34. [5] Surg Obes Relat Dis 2015;11:222—8. Hackert T, Sachsenmaier M, Hinz U, et al. Locally advanced pancreatic cancer: neoadjuvant therapy with folfirinox results in resectability in 60% of the patients. Ann Surg 2016;264:457—63. http://dx.doi.org/10.1097/SLA. 0000000000001850
Objective For patients with locally advanced and unresectable pancreatic cancer (PDAC), neodadjuvant treatment and consecutive surgical exploration have been studied during the last decade with various neoadjuvant therapies including chemotherapy and combinations with radiation. Aim of the study was the evaluation of neoadjuvant therapy with a focus on Folfirinox. Methods All consecutive patients undergoing surgery for PDAC after neoadjuvant treatment were analyzed (clinicopathological characteristics, secondary resection rates, outcome). Patients receiving Folfirinox were compared with other treatment regimens. Results Between December 2001 and June 2015, 575 patients received neoadjuvant treatment and were scheduled for resection after re-staging. A successful resection was achieved in 292 patients (50.8%). Resection rates following Folfirinox were 61% (76/125 patients) compared with 46% (150/322 patients) after gemcitabine and radiation, and 52% (66/128 patients) after other treatments (P = 0.026). Median overall survival was 15.3 months after resection vs 8.5 months after exploration alone (P < 0.0001). Subgroup median survival was 16.0 months (Folfirinox) vs 16.5 months (gemcitabine) and 14.5 months (others) with 3-year survival of 28.1%, 23.2%, and 19.7%, respectively (P = 0.8582). By multivariable analysis, Folfirinox was confirmed to be independently associated with a favorable prognosis. Conclusions Folfirinox is a valuable treatment option in the neoadjuvant therapy of PDAC. From the present data, which represent the largest available study population to date, Folfirinox seems to be the most effective protocol resulting in a significantly better secondary resection rate and overall survival than other treatments. It should be considered in all patients fit for this regimen and consecutive surgical exploration.
472 Comments 1. In spite of improvement in medical therapy, surgery remains the only treatment with curative intent for pancreatic cancer. Because of vascular or adjacent organ involvement, surgical possibilities are limited for PDAC. The authors suggest that neo-adjuvant chemotherapy, and in particular, Folfirinox, can obtain a 61% resectability rate and increase overall survival. 2. As of today, there are no randomized studies available concerning the benefits of Folfirinox in PDAC, but several trials are underway, notably one in France. One recent meta-analysis found that secondary resection was possible in only 43% of patients after Folfirinox. 3. Of note, even if these authors highlight a high surgical resection rate, the R0 resection rate was only 40.8% after Folfirinox (vs. 31.3% after gemcitabine and 27.3% after other treatment options, P = 0.041), this low rate being responsible for the moderate benefit in overall survival.
C. Mariette, S. Benoist 4. The two groups re not comparable as concerned the study periods under analysis as well as age, both being substantial causes of bias. Cancer-specific survival analysis would have been more pertinent. 5. This study has several potential biases that could have influenced outcome, notably (i) absence of precisions concerning the number of patients included in the treatment plan who, after that period, were still not amenable to surgery, (ii) a highly probable mix of borderline and locally advanced tumors, (iii) a selection bias, (iv) a shorter follow-up in the Folfirinox group (only recently available in the therapeutic armamentarium, lessening the chances that recurrence or death could occur), (v) survival was calculated starting with the date of surgery (starting with the beginning of treatment would have been more correct), (vi) 51 patients with initial metastatic disease were included. Déclaration de liens d’intérêts Les auteurs déclarent ne pas avoir de liens d’intérêts.