Séances ordinaires

Résumé
Nous rapportons un cas rare de fistule œso-trachéale (FOT) post-traumatique, opérée par bi-exclusion œsophagienne avec plastie digestive. Les causes de fistule œso-trachéale sont multiples. Les fistules œso-trachéales post-traumatiques consécutives à un traumatisme « fermé » du thorax sont rares, mais graves. Les principaux symptômes sont : dysphagie, fausse route alimentaire, renvois d’aliments ingérés mélangés de sécrétions oro-bronchiques par la bouche, amaigrissement et infections broncho-pulmonaires par inhalation. Le diagnostic dans ce contexte, est souvent méconnu au stade initial. Les examens morphologiques diagnostiques sont : fibroscopie digestive haute, fibroscopie bronchique, transit radio-opaque de l’œsophage et scanner thoracique. La mortalité en l’absence d’intervention est de 80 % versus, 9,3 % en cas d’intervention. La mortalité est due aux complications médiastinales et pulmonaires. Le traitement de ce type de fistule œso-trachéale est uniquement chirurgical. Les indications thérapeutiques sont fonction de taille de la fistule. En cas de fistule géante, la bi-exclusion de l’œsophage sus et sous fistuleux avec plastie digestive, soit par l’estomac soit par le colon, est une solution intéressante. Ce procédé est un moyen thérapeutique efficace, adapté à un pays en voie de développement comme le nôtre, où les conditions socio-économiques et médicales sont encore limitées.

Abstract
We report an interesting and rare case of post traumatic œso-tracheal fistula operated by esophageal bi-exclusion and digestive plasty. The post traumatic œso-tracheal fistula is a rare but grave pathology. Syndrome includes: dysphagia, food choking, spilling out food with phlegm, weights lost and other complications in relation with swallowing problems. The diagnosis is often not made at the initial period. Clinical diagnosis: oeso-gastroscopy, bronchoscopy, esophageal opacification transit and chest CT scan, œso-gastroscopy, bronchoscopy, esophageal opacification transit are good methods of examination and diagnose even the size of fistula is small. Death can be caused by lung and other mediastinal complications. Without surgery, the mortality rate is 80%, but it decreases to 9.3% in case of surgery. Treatment by medicines, biological glue, tracheal and oesophageal prosthesis are only temporarily effective. Surgical treatment is the unique effective possibility for post-traumatic œso-tracheal fistula; it depends of the dimension of fistula. In case of large fistula, the bi-exclusion method of œsophagus with digestive replacement by the stomach or the colon is an effective solution, especially in low and developed economy country as Cambodia.

Abstract
Objective: We analyzed the impact of chemotherapy-related liver injuries (CALI) and pathological tumor regression grade (TRG) on long-term prognosis in patients undergoing liver resection for colorectal metastases (CRLM).

Background: CALI worsen the short-term outcomes of liver resection, but their impact on prognosis is unknown. Recently, a prognostic role of TRG has been suggested.

Methods: Patients undergoing liver resection for CRLM between 1998 and 2011 and treated with oxaliplatin and/or irinotecan-based pre-operative chemotherapy were eligible for the study. Patients with operative mortality or incomplete resection (R2) were excluded. All specimens were reviewed to assess CALI and TRG.

Results: 323 patients were included. Grade 2-3 sinusoidal obstruction syndrome (SOS) was present in 124 patients (38.4%), nodular regenerative hyperplasia in 63 (19.5%), grade 2-3 steatosis in 73 (22.6%), and steatohepatitis in 30 (9.3%). Among all patients, 22.9% had TRG1-2 (major response), 21.4% had TRG3 (partial response), and 55.7% had TRG4-5 (no response).

The higher the SOS grade the lower the pathological response: TRG1-2 and TRG4-5 occurred in 16.9% and 64.5% of patients with grade 2-3 SOS vs. 26.6% and 50.3% of patients with grade 0-1 SOS, respectively (p=0.036).

After a median follow-up of 36.9 months, 5-year survival was 38.6%. CALI did not negatively impact survival. Multivariate analysis showed that grade 2-3 steatosis was associated with better survival than grade 0-1 steatosis (5-year survival 52.5% vs. 35.2%, p=0.003). TRG stratified patient prognosis: 5-year survival 60.4% in TRG1-2, 40.2% in TRG3, and 29.8% in TRG4-5 (p=0.007).

Microvascular and biliary invasion were present in 37.8% and 5.6% of patients and negatively impacted outcome (5-year survival 23.3% vs. 45.7% if absent [p=0.024] and 0% vs. 42.1% [p=0.026], respectively).

Conclusions: CALI do not negatively impact long-term prognosis, but the tumor response is reduced in patients with grade 2-3 SOS. TRG was confirmed to be a crucial prognostic determinant, overcoming traditional morphological parameters. Steatosis was found to have a protective effect on survival. Identification of microvascular and biliary invasion contributed to prognosis assessment.

Résumé
Notre expérience dans l’hyperparathyroïdie HPT a commencé en 1975 avec le traitement de l’HPT secondaire. Depuis cette date jusqu'en 2012 nous avons opéré 1 531 patients. Nous avons divisé cette expérience en deux périodes, la première de 1975 à 1995 (277 patients) et la seconde de 1996 à 2012 (1 254 patients). Le nombre élevé de patients adressés au chirurgien associé aux nouvelles techniques (vidéoscopie, chirurgie mini invasive, PTH peropératoire) nous ont conduits à des réflexions sur l'approche chirurgicale dominés par les examens de localisation préopératoires et l’utilisation de la PTH peropératoire. Nous avons observé une corrélation parfaite entre l'échographie et la scintigraphie, la présence de formes multiglandulaires dans 12,7 % des cas. Selon nous, la baisse du taux de la PTH n'a de valeur que si elle s'associe à un retour à la normalité. En conclusion, en accord avec plusieurs publications récentes, nous estimons qu'une réflexion d'ensemble doit porter sur la meilleure approche pour offrir aux patients la meilleure chance d'une guérison définitive. Une exploration de toutes les glandes associées à une meilleure interprétation des résultats de la PTH peropératoire (retour à la normalité ou une chute progressive après encore 10 minutes) semble être la meilleure attitude.

Abstract
Our experience regarding HPT began in 1975 with the treatment of secondary HPT. Now the interventions for secondary and tertiary HPT are more than 900. In this paper we only treat primitive HPT, 1531 patients operated until the end of 2012. This experience was divided into two periods, the first from 1975 to 1995 (277 patients) and the second from 1996 to 2012 (1254 patients); this allows us to make some observations. The most important is the use of intraoperative PTH. Even when there is a perfect correlation between ultrasound and scintigraphy, the MGD is present in 12.7% of cases in our experience. But the most interesting is the decrease of PTH. In 5.1% of these cases, a decline of more than 50% was obtained, but not in the normal range, so we explore the other side of the neck through the same incision and a MGD was present. We did not find MGD when the PTH was in the normal range. In conclusion, in agreement with several recent papers (Moalem, Mc Gill, Norman, Mazzaglia, Harari, Slepavicius, Nagar, Yavuz) on the surgical treatment of HPT, we believe it should be made a reflection on what constitutes the best approach for this condition in order to offer to the patients the best chance of a definitive cure. A reassessment of the exploration of all the glands associated with a better interpretation of the results of the intraoperative PTH (return to normal or a progressive fall after another 10 minutes) in our opinion seems to be the best attitude.

Résumé
En novembre 2011 arriva au secours de nôtre hôpital une patiente de 64 ans, présentant un tableau clinique, laboratoire et d'imagerie d’infarctus intestinal, la symptomatologie étant débuté 18 heures avant.
La patiente ne souffrait d’aucune pathologie emboligène. Antécédents : mastectomie pour carcinome ductal de premier stade, traitement avec tamoxifène puis exémestane, parathyroïdectomie pour iperparathyroïdisme primaire, diagnose endoscopique et bioptique de colite ischémique ou maladie inflammatoire intéstinale.
La patiente fut opérée en urgence, l'exploration révélant multiples et délimitées zones nécrotiques du grêle distal et du côlon. À cause de la rapide et progressive aggravation hémodynamique, due au choc septique, on n'a pu réaliser qu’une iléostomie terminale décompressive. Après défonctionnalisation de l'intestin ischémique et nécrotique, rééquilibration hydroélectrolytique, antibiotiques, traitement de l'acidose métabolique, il était devenu possible réaliser un second look chirurgical pour exclure la progression de l'ischémie et faire une résection iléocolique jusqu'au côlon gauche, avec sigmoïde occlus et iléostomie terminale. Trois mois et demi après la continuité intestinale a été rétablie par anastomose iléosigmoïdienne non protégée, résultant en rapide guérison.
En considérant l'absence de maladie emboligène et la multifocalité des infarctus, la cause était vraisemblablement l'occlusion thrombotique de artères collatérales de petit et moyen diamètre de l'artère mésentérique supérieure. L'étiologie vasculitique avait été exclu par l'anatomopathologie et par l'absence de manifestations auto-immunes soit cliniques soit de laboratoire.
Nous avons donc formulé l'hypothèse étiologique de sténose artérielle thrombotique due au traitement avec tamoxifène (pendant 3.5 ans), déclenchée par le changement thérapeutique avec exémestane (8 mois avant). Il est bien connu que ces médicaments provoquent une augmentation du risque thromboembolique, mais référée uniquement à l'ischémie cérébrovasculaire et cardiaque. Il n'a été signalé en littérature jusqu'à présent aucun cas d’ischémie mésentérique liée à ces médicaments.

Abstract
In November 2011 a 64 years old woman came to our emergency room, presenting clinical, blood and consistent imaging features with intestinal infarction, the symptoms having started 18 hours before.
The patient did not show any emboligen disease. In her medical history: mastectomy for an early stage breast cancer, treatment with tamoxifen then exemestane, right hip replacement, endoscopical and bioptic diagnosis of ischemic colitis or inflammatory bowel disease.
The patient underwent an emergent laparotomy, which revealed multiple and confined necrotic patches in her terminal ileum and colon. Because of fast and progressive hemodynamic worsening, due to onset of septic shock, only a diverting ileostomy could be performed. After defunctioning ischemic and necrotic bowel, restoring intravascular volume, giving wide spectrum antibiotics, correcting metabolic acidosis and having reached a stable hemodynamic balance, it became possible to carry out the second surgical look. Extension of ischemia could be excluded and an ileocolic resection, including the descending colon, was performed, leaving the sutured sigmoid stump in the pelvis and the ileostomy previously done. Three and a half months later bowel transit has been restored by an ileum-sigmoid anastomosis, followed by a fast recovery.
Because of absence of emboligen diseases and considering the patchy necrotic lesions, the cause is likely to be a thromboembolic occlusion of medium and small arterial branches of the superior and inferior mesenteric arteries. The vasculitic etiology has been excluded by pathological findings, lack of clinical symptom and by blood immunological tests.
Hence we hypothesized a thrombotic stenosis, due to the long lasting treatment with tamoxifen (3.5 years), critically deteriorated by the therapeutical switch with exemestane (8 months before), as the etiologic factors.
It is well known indeed as those drugs can rise the thromboembolic risk, but it is referred only to cerebrovascular and cardiac ischemia. Until the date, there are no reports about mesenteric ischemia related to neither tamoxifen nor exemestane.

Abstract
Objective: to evaluate the role of acinar cells density, pancreatic fatty infiltration and fibrosis in the developement of major complications after pancreaticoduodenectomy (PD).

Material and Methods: we prospectively collected a series of 46 consecutive PD performed in our center. The variables considered were: age, gender, ASA score, pancreatic texture, Wirsung diameter, postoperative mortality and morbility (pancreatic fistula - PF, post pancreatectomy haemorrage - PPH, delayed gastric emptying - DGE), hospital stay, histology and preoperative biliary drainage. Both univariate and multivariate models were used to determine factors correlated to PF. Analisys regarding histological findings (pancreatic fibrosis, pancreatic fatty infiltration, acinar cell frequency) and correlation with PF was conducted. Statistical analysis was performed using SPSS v10.0. Differences between categorical variables were evaluated by X2 analysis, while Student’s T test was used for all comparison among continuous variables. Univariate and Multivariate analysis were performed using SPSS v 10.0.

Results: most of the patients were resected for malignant diseases (19 pancreatic head carcinoma, 2 distal biliary tract carcinoma, 18 ampullary malignancy, 2 neuroendocrine tumors, 1 duodenal cancer, 1 degenerated IPMN, 3 non malignant diseases). All patients underwent standard PD with dunking, sutureless, end to end PJ anastomosis with distal gastrectomy and end to side hepatico-jejunostomy. Overall morbility rate and operative mortality rate were respectively 43,2 and 5 %. PF rate was 21,7 %. PPH rate was 17,8 % and DGE 27,3 %. At univariate analysis PF rate correlates with Wirsung diameter < 3 mm (p=0,015 OR: 6,25), pancreatic cancer (p=0,016 OR = 0,37), soft pancreatic texture ( p=0,001 OR= 0,37), male gender (p=0,01). 45,5% of high fatty pancreas patients developed PF; 88,9% of patients with PF had high frequency of acinar cells; patients with fatty or high fatty pancreas, in more than 70 % of cases had high acinar cells density. No correlation was found between BMI and the severity of fatty infiltration score and between BMI > 25 and PF.

Conclusions: In our series PF fistula rate correlates with pancreatic texture and wirsung diameter. BMI was not usefull to identify PF high risk patients and it has not been shown to be an accurate index of pancreatic fatty infiltration. High acinar cells frequency on pancreatic cut edge could be an interesting marker to identify high risk pancreas.

Intervenant : P BACHELIER (Strasbourg)

Abstract
Background: liver packing (LP) is an accepted technique to achieve rapid control of hemostasis after liver trauma requiring emergent laparotomy. Little is known about the indications and outcomes of this technique used in the context of massive bleeding during orthotopic liver transplantation (OLT)

Aim: to investigate the characteristics and outcomes of patients treated by LP during OLT

Patients and methods: recipients of a OLT treated by LP for hemorrhage in the period from 2003 to 2013 were identified from operating theater reports. Clinical charts were retrospectively reviewed to collect demographic data, patients and donor characteristics at transplantation, operatory variables, the indication for LP and postoperative outcomes.

Results: LP was performed in 98/1399 (7%) of patients, 27% of whom recipients of a second OLT. Median age, Model for End-Stage Liver Disease score and body mass index was 54 years, 18 and 23.7 kg/m2, respectively. At the moment of packing, median pH, lactate and temperature was 7.24, 4.8 mEq/L and 34.6 °C, respectively. A mean of 7,860 ml of blood and 8,280 ml of plasma was transfused during the operation. LP was invariably effective in achieving hemostasis. Depacking was performed during a second (92%) or third (8%) operation, for a mean open abdomen duration of 3 days. Postoperative patient and graft survival rate was 90% and 84%, respectively. Severe postoperative morbidity (Dindo-Clavien 3 – 4) was observed in 26% of the patients.

Conclusions: Despite significant postoperative morbidity, LP is effective in achieving hemostasis during OLT, with acceptable patient and graft survival.

Intervenant : D CHERQUI (Paris)