Nicolas Briez

University of Lille Nord de France, Lille, Nord-Pas-de-Calais, France

Are you Nicolas Briez?

Claim your profile

Publications (22)86.21 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: Despite the prevalence of complex ventral hernias, there is little agreement on the most appropriate technique or prosthetic to repair these defects, especially in contaminated fields. Our objective was to determine French surgical practice patterns among academic surgeons in complex ventral hernia repair (CVHR) with regard to indications, most appropriate techniques, choice of prosthesis, and experience with complications. A survey consisting of 21 questions and 6 case-scenarios was e-mailed to French practicing academic surgeons performing CVHR, representing all French University Hospitals. Forty over 54 surgeons (74%) responded to the survey, representing 29 French University Hospitals. Regarding the techniques used for CVHR, primary closure without reinforcement was provided in 31.6% of cases, primary closure using the component separation technique without mesh use in 43.7% of cases, mesh positioned as a bridge in 16.5% of cases, size reduction of the defect by using aponeurotomy incisions without mesh use in 8.2% of cases. Among the 40 respondents, 36 had experience with biologic mesh. There was a strong consensus among surveyed surgeons for not using synthetic mesh in contaminated or dirty fields (100%), but for using it in clean settings (100%). There was also a strong consensus between respondents for using biologic mesh in contaminated (82.5%) or infected (77.5%) fields and for not using it in clean setting (95%). In clean-contaminated surgery, there was no consensus for defining the optimal therapeutic strategy in CVHR. Infection was the most common complication reported after biologic mesh used (58%). The most commonly reported influences for the use of biologic grafts included literature, conferences and discussion with colleagues (85.0%), personal experience (45.0%) and cost (40.0%). Despite a lack of level I evidence, biologic meshes are being used by 90% of surveyed surgeons for CVHR. Importantly, there was a strong consensus for using them in contaminated or infected fields and for not using them in clean setting. To better guide surgeons, prospective, randomized trials should be undertaken to evaluate the short- and long-term outcomes associated with these materials in various surgical wound classifications.
    Journal of Visceral Surgery 12/2013; · 1.17 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To compare the outcomes of a strategy of surveillance versus surgical resection in patients with esophageal cancer (EC) experiencing complete clinical response (cCR) after chemoradiation (CRT). In EC, it remains unclear whether a strategy of surveillance or esophagectomy is appropriate after cCR to CRT. From 1995 to 2009, 222 operable patients had a cCR based on the results of a computed tomographic scan, endoscopy with biopsies and, when performed, a positron emission tomographic scan. Through an intention-to-treat case-control study, 59 patients treated with CRT and surveillance (group Surv) were matched 1:2 with 118 patients treated by CRT followed by surgery (group Surg), according to age, gender, tumor location and stage, histology, American Society of Anesthesiologists score, and nutritional status. The 2 groups were comparable according to the matched variables (P > 0.276). In group Surg, the postoperative mortality rate was 4.2% with evidence of residual tumor in 34.6% of specimens. In group Surv, 2 salvage esophagectomies were performed. Despite the higher dose of radiotherapy received in group Surv (50 vs 45 Gys, P = 0.003), median survival was lower (31 vs 83 months, P = 0.001), with disease recurrence that was more frequent (50.8% vs 32.7%, P = 0.021), occurred earlier (7.8 vs 19.0 months, P = 0.002) and more often locoregional (46.7% vs 16.2%, P = 0.007) in nature. Surgical resection was independently associated with less recurrence [odds ratio = 0.4, 95% confidence interval (CI): 0.2-0.8, P = 0.006] and better survival (hazard ratio = 0.5, 95% CI: 0.3-0.8, P = 0.006). Survival of EC patients with a cCR after CRT is better after surgery compared to simply surveillance. In patients of low operative risk and operable disease, surgery should be considered to improve control of locoregional disease and to overcome the inherent limitations of clinical response assessment.
    Annals of surgery 10/2013; · 7.90 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The mechanisms of esophageal reflux leading to esophageal adenocarcinoma (EA) remain poorly understood. This study appraises critically an operatively induced chronic reflux rat model. We randomized 108 Sprague-Dawley rats into 2 experimental groups; one was performing esophagoduodenal (ED) anastomosis with or without gastrectomy to induce duodeno-esophageal reflux (DER group; n = 63), and the other involved duodeno-gastro-esophageal reflux (DGER group; n = 45). Control groups included (i) Roux-en-Y esophagojejunal anastomosis, (ii) laparotomy alone, (iii) subtotal gastrectomy to induce duodenogastric reflux (DGR group), and (iv) the same procedure as in the DGER group plus proton pump inhibition (PPI group). The esophagus underwent histologic and molecular analyses. The prevalence of Barrett's esophagus (BE), dysplasia, and EA in the experimental groups was 41%, 7%, and 11%, respectively. Histologic and molecular analyses in groups DER, DGER, and DGR suggested that BE occurred through de novo intestinal metaplasia and proximal migration of duodenal cells. No distant metastases were identified. The molecular characteristics of both BE and EA were similar to humans. BE was more common, and dysplasia and EA less frequent in the DER group when compared with the DGER group (44% vs 24% [P = .038] and 7% vs 25% [P = .012], respectively). Compared with the DGER group, carcinogenic sequence occurred less frequently in the PPI-treated group (P = .019). Despite pathophysiologic differences with humans, the rat model of esophagoduodenostomy reproduces accurately histologic and molecular lesions in the carcinogenetic sequence of BE and allowed us to identify novel, tumor-associated proteins that may be potential biomarkers and new therapeutic targets in EA.
    Surgery 09/2013; · 3.37 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: BACKGROUND: In infected incisional ventral hernias (IVHs), the use of a synthetic non-absorbable mesh is not recommended and biological meshes hold promise. However, the level of evidence for their safety and efficacy remains low. METHODS: The SIMBIOSE trial is a multicenter, phase III, randomized, controlled trial comparing the use of a biological mesh versus traditional wound care in patients with an IVH. The primary end point is 6-month infectious and/or wound morbidity. Secondary end points are wound infection and recurrent hernia rates, post-operative pain, quality of life, time to heal, reoperation need, impact of the cross-linked mesh structure, and a medico-economic evaluation. One hundred patients need to be included. RESULTS: The main results expected with biological mesh use are a significant decrease of post-operative morbidity, hernia recurrence, time to heal, and costs with an improved quality of life. CONCLUSIONS: For the first time, the impact of biological meshes in the treatment of IVHs will be evaluated in an academic, randomized, phase III trial to provide scientific evidence (NCT01594450).
    Trials 05/2013; 14(1):131. · 2.21 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Morbidity after oesophageal cancer surgery remains high, mainly due to major postoperative pulmonary complications (MPPCs). The aim of this study was to test the hypothesis that hybrid minimally invasive oesophagectomy (HMIO) decreases the 30-day MPPC rate without compromising oncological outcomes. Consecutive patients undergoing curative oesophagectomy for cancer by laparoscopic gastric mobilization and open thoracotomy (HMIO) between January 2004 and December 2009 were matched to randomly selected patients undergoing a totally open approach during the same study interval. Matching variables were age, sex, cancer stage, location of the primary tumour, histological subtype, American Society of Anesthesiologists grade, malnutrition, neoadjuvant chemoradiation and epidural analgesia. MPPCs at 30 days were significantly less frequent after HMIO compared with open surgery (15·7 versus 42·9 per cent; P < 0·001). Postoperative in-hospital mortality and overall morbidity rates were 4·3 and 47·5 per cent respectively, again significantly lower in the HMIO group: 1·4 versus 7·1 per cent (P = 0·018) and 35·7 versus 59·3 per cent (P < 0·001). In multivariable analysis, HMIO, adenocarcinoma subtype, epidural analgesia and surgery after 2006 were independent protective factors against MPPCs, and HMIO was independently protective against acute respiratory distress syndrome (ARDS). Lymph node yields and survival were similar in the two groups. HMIO for oesophageal cancer, using laparoscopic gastric mobilization and open right thoracotomy, offered a substantial and independent protective effect against MPPCs, including ARDS, without compromising oncological outcomes. Copyright © 2012 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd.
    British Journal of Surgery 11/2012; 99(11):1547-53. · 4.84 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: BACKGROUND: Percutaneous endoscopic gastrostomy is not widely used in malnourished esophageal cancer (EC) patients because of concerns about its feasibility in frequently obstructive tumors, suitability of the stomach as an esophageal substitute, and potential for metastatic inoculation. A percutaneous radiological gastrostomy (PRG) could be an optimal alternative. METHODS: Experience with PRG among 1,205 consecutive patients presenting with EC from 2002 to 2011 in our department was retrospectively reviewed. PRG was mostly utilized for malnourished patients for whom neoadjuvant chemoradiation was scheduled. The rates of both successful placement and major related complications (Dindo-Clavien ≥III) were analyzed. A matched cohort analysis was constructed in patients who underwent esophagectomy with gastroplasty (n = 688) to evaluate the impact of PRG placement on the suitability of the gastric conduit and on postoperative course. For 78 resected patients with PRG (PRG group), 156 randomly selected controls without PRG (no PRG group) were matched 2:1 for gender, age, ASA grade, clinical TNM stage, and neoadjuvant treatment delivery. RESULTS: PRG placement was planned in 269 (22.3 %) patients mainly with locally advanced EC (63.8 %). PRG placement was feasible in 259 (96.3 %) patients. Sixty-day PRG-related mortality and major morbidity rates were 0 and 3.8 % respectively. For resected patients, the PRG and no PRG groups were comparable regarding perioperative characteristics, except for malnutrition, which was more frequent in the PRG group (P < 0.001). At the time of operation, PRG takedown and site closure were uncomplicated and the use of the stomach was possible in all 78 patients. Despite a higher malnutrition rate at presentation in the PRG group, rates of overall morbidity, and morbidity related to esophageal surgery, were similar between the two groups (P > 0.258). CONCLUSION: PRG is feasible, safe, and useful in nonselected patients with EC and does not compromise the suitability of the stomach as an esophageal substitute in patients deemed to be resectable.
    Surgical Endoscopy 09/2012; · 3.43 Impact Factor
  • N. Briez, C. Mariette
    Journal de Chirurgie Viscérale. 02/2012; 149(1):85–90.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Open transthoracic oesophagectomy is the standard treatment for infracarinal resectable oesophageal carcinomas, although it is associated with high mortality and morbidity rates of 2 to 10% and 30 to 50%, respectively, for both the abdominal and thoracic approaches. The worldwide popularity of laparoscopic techniques is based on promising results, including lower postoperative morbidity rates, which are related to the reduced postoperative trauma. We hypothesise that the laparoscopic abdominal approach (laparoscopic gastric mobilisation) in oesophageal cancer surgery will decrease the major postoperative complication rate due to the reduced surgical trauma. The MIRO trial is an open, controlled, prospective, randomised multicentre phase III trial. Patients in study arm A will receive laparoscopic-assisted oesophagectomy, i.e., a transthoracic oesophagectomy with two-field lymphadenectomy and laparoscopic gastric mobilisation. Patients in study arm B will receive the same procedure, but with the conventional open abdominal approach. The primary objective of the study is to evaluate the major postoperative 30-day morbidity. Secondary objectives are to assess the overall 30-day morbidity, 30-day mortality, 30-day pulmonary morbidity, disease-free survival, overall survival as well as quality of life and to perform medico-economic analysis. A total of 200 patients will be enrolled, and two safety analyses will be performed using 25 and 50 patients included in arm A. Postoperative morbidity remains high after oesophageal cancer surgery, especially due to major pulmonary complications, which are responsible for 50% of the postoperative deaths. This study represents the first randomised controlled phase III trial to evaluate the benefits of the minimally invasive approach with respect to the postoperative course and oncological outcomes in oesophageal cancer surgery. NCT00937456 (ClinicalTrials.gov).
    BMC Cancer 01/2011; 11:310. · 3.33 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Gastric and oesophageal cancers are among the leading causes of cancer-related death worldwide. By contrast with the decreasing prevalence of gastric cancer, incidence and prevalence of oesophagogastric junction adenocarcinoma (OGJA) are rising rapidly in developed countries. We provide an update about treatment strategies for resectable OGJA. Here we review findings from the latest randomised trials and meta-analyses, and propose guidelines regarding endoscopic, surgical, and perioperative treatments. Through a team approach, members from all diagnostic and therapeutic disciplines, such as gastroenterologists, surgeons, oncologists, radiologists, and radiotherapists, can effectively administer a range of treatment modalities.
    The Lancet Oncology 11/2010; 12(3):296-305. · 25.12 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: But du travail Alors que la douleur chronique est bien décrite après cure de hernie inguinale, les données après cure d’éventration sont rares. Le but de cette étude était d’évaluer l’incidence, les caractéristiques et les facteurs prédictifs de la douleur chronique après cure d’éventration. Méthodes Notre étude rétrospective a inclus de 2005 à 2008 tous les patients opérés d’une cure d’éventration avec plaque intrapéritonéale biface. L’évaluation de la douleur chronique a été réalisée par un observateur indépendant, au moyen de 4 outils validés : (I) une échelle numérique en 11 points (score supérieur à 3 : douleur chronique, supérieur à 7 : douleur sévère) ; (II) le questionnaire de la douleur de saint Antoine évaluant les composantes sensorielle et affective de la douleur ; (III) un questionnaire de douleur neuropathique, (iv) une échelle numérique en 5 points de satisfaction globale. Résultats Après un suivi médian de 21,1 mois, 134 patients ont été évalués (109 ont complété les questionnaires, soit 81,3 %).Le taux de douleur chronique était de 28,4 % (soit 31 patients) dont 6,6 % de douleur sévère (7 patients). Les composantes de la douleur étaient sensorielle pour 19,3 % (6/31 patients) ou affective 45,1 % (14 /31patients). Une douleur neuropathique était présente pour 6,4 % (7 patients). Le score de satisfaction global était supérieur ou égal à 3/4 pour 82,3 % (90 patients). En analyse multivariée, les facteurs prédictifs de douleur chronique étaient la toux chronique (p = 0,001, OR 11,4), un abcès pariétal postopératoire (p = 0,034, OR 4,5) et une récidive d’éventration (p = 0,05, OR 3,8). Conclusion La douleur chronique après cure d’éventration est fréquente mais rarement sévère, principalement liée à la composante affective. La satisfaction globale des patients reste bonne. Une évaluation à plus long terme est nécessaire.
    Ultrasound in Medicine and Biology - ULTRASOUND MED BIOL. 01/2010; 147(4):13-14.
  • [Show abstract] [Hide abstract]
    ABSTRACT: But du travail La morbidité de l’œsophagectomie pour cancer reste élevée, en particulier du fait des complications pulmonaires. L’essor de la coelioscopie est essentiellement fondé sur la perspective d’une moindre morbidité postopératoire par diminution de l’agression liée à la voie d’abord. Le but de notre travail était de comparer la morbidité pulmonaire à 30 jours des patients opérés d’une œsophagectomie deux voies (Lewis Santy) pour cancer coelio-assistée (gastrolyse coelioscopique, groupe C) versus voie ouverte standard (groupe O). Patients et Méthodes De janvier 2002 à juin 2008, 331 œsophagectomies pour cancer ont été réalisées. La technique de gastrolyse coelioscopique a été introduite progressivement depuis 2005 pour des patients non sélectionnés. Dans cette étude prospective comparative, 67 patients consécutifs du groupe C ont été appariés à 183 patients témoins du groupe O selon les critères d’âge, de sexe, de localisation et de stade tumoral, de type histologique, de score ASA, de dénutrition et d’administration d’une radiochimiothérapie néo-adjuvante. Résultats Les deux groupes étaient comparables sur les critères d’appariement retenus. Les taux de mortalité et de morbidité globale postopératoires étaient de 2,4 % et 41,2 %, respectivement. Le taux de conversion était de 1,6 % (n = 4). Les complications pulmonaires étaient significativement moins fréquentes dans le groupe C (16,4 % vs 34,4 %, p = 0,006) et avec moins de complications pulmonaires majeures (14,9 % vs 30,0 %, p = 0,016). En comparant les groupes C et O, il n’existait pas de différence en termes de mortalité (1,5 % vs 2,7 %), de morbidité globale (38,8 % vs 42,0 %), de fistule anastomotique (5,9 % vs 3,8 %), de réintervention (6,0 % vs 8,7 %), de dilatation de plastie (5,9 % vs 2,2 %), ni de complication septique (13,4 % vs 18,6 %). Conclusion Il s’agit de la plus importante étude comparative prospective montrant une diminution de l’incidence et de la gravité des complications pulmonaires grâce à l’approche mini-invasive dans le cancer de l’œsophage. Les résultats carcinologiques au long cours devront être validés. Un essai randomisé multicentrique français débute en ce sens.
    Ultrasound in Medicine and Biology - ULTRASOUND MED BIOL. 01/2010; 147(4):32-32.
  • [Show abstract] [Hide abstract]
    ABSTRACT: But du travail Les stratégies de radiochimiothérapie (RCT) exclusive et de RCT néo-adjuvante sont deux options de traitement à visée curative du cancer de l’œsophage. Le but du travail était de comparer les survies des patients en réponse complète après RCT exclusive (groupe RCTE) versus après RCT suivie de chirurgie (groupe RCTC). Patients et Méthodes De 1995 à 2007, parmi 1176 patients traités pour cancer de l’œsophage, une étude cas témoins en intention de traiter a été réalisée à partir de 55 patients du groupe RCTE appareillés à 111 patients témoins du groupe RCTC selon les critères d’âge, de sexe, de localisation et de stade tumoral, de score ASA, de type histologique et de dénutrition. La réponse à la RCT était évaluée par fibroscopie+ biopsies et tomodensitométrie dans le groupe RCTE et l’anatomopathologie sur la tumeur primitive dans le groupe RCTC. Résultats Les deux groupes étaient comparables avec, dans les groupes RCTE et RCTC respectivement, un âge moyen de 59 vs 57 ans, un taux de carcinomes épidermoïdes de 87 % vs 91 %, un score ASA 2 ou 3 de 65 % vs 76 %, une localisation tumorale au tiers moyen ou inférieur de l’œsophage dans 71 % vs 77 % des cas et une tumeur avancée dans 64 % vs 60 % des cas. Les délais moyens d’apparition de la récidive étaient respectivement de 10 vs 18 mois (p = 0,012). Les médianes et taux de survie à 5 ans étaient respectivement de 25 vs 61 mois et 19 % vs 50 % (p = 0,001). Conclusion La survie des patients en réponse complète est meilleure après RCT suivie de chirurgie qu’après RCT exclusive. Il convient donc d’opérer si possible ces patients, pour un meilleur contrôle locorégional de la maladie mais surtout parce que l’évaluation de la réponse complète morphologique apparaît ineffective.
    Ultrasound in Medicine and Biology - ULTRASOUND MED BIOL. 01/2010; 147(4):20-21.
  • [Show abstract] [Hide abstract]
    ABSTRACT: To study the expression of MUC1 and MUC4 mucins in Barrett-associated oesophageal adenocarcinoma and coexisting lesions of the carcinogenic sequence (normal mucosa, metaplasia, dysplasia) if present, and to investigate their prognostic significance. The expression profiles of MUC1 and MUC4 were investigated by immunohistochemistry in tissue samples obtained from consecutive patients with primary surgically resected lower third oesophageal adenocarcinoma (OA) between 1997 and 2002. Histopathological parameters, recurrence and long-term survival were correlated with the number of cells stained. All 52 patients exhibited OA, with 25 patients (48.1%) having associated Barrett oesophagus lesions (metaplasia or/and dysplasia). MUC1 and MUC4 were expressed in 52 and 41 of the 52 patients with adenocarcinoma (100% and 78%), respectively. All samples expressed MUC1 strongly. The prevalence of MUC4 staining was significantly decreased in metaplasia compared with normal mucosa (53% versus 92%, p<0.001). No correlation was found between the level of MUC1 or MUC4 expression in OA and histopathological variables, recurrence or survival. MUC1 and MUC4 are strongly expressed in OA. The results do not support a role for these two membrane-bound mucins as either a phenotypic or a prognostic marker for the development of Barrett OA. There are several other membrane-bound mucins that have not yet been evaluated in this situation.
    Journal of clinical pathology 12/2009; 62(12):1144-6. · 2.43 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The hypoxia inducible factor (HIF) signaling pathway is known as the main renal carcinogenetic pathway. MUC1, an O-glycoprotein membrane-bound mucin, is overexpressed in clear renal cell carcinomas (cRCC) with correlation to two major prognostic factors: tumor-node-metastasis stage and nuclear Fürhman grade. We questioned whether there is a direct link between the HIF pathway and MUC1 overexpression in renal tumors. Interestingly, we observed concomitant increase of HIF-1alpha and MUC1 in metastatic cRCC group versus nonmetastatic cRCC group. Using different renal cell models and small interfering RNA assays targeting either HIF-1alpha or YC-1, a HIF-1 pharmacologic inhibitor, we showed induction of MUC1 expression under hypoxia by a HIF-dependent mechanism. Chromatin immunoprecipitation assay showed a direct binding of HIF-1alpha at the MUC1 promoter. In addition, combined site-directed mutagenesis and gel shift assay allowed the identification of two functional putative hypoxia responsive elements at -1488/-1485 and at -1510/-1507 in the promoter. Using a rat kidney model of ischemia/reperfusion, we confirmed in vivo that clamping renal pedicle for 1 hour followed by 2 hours of reperfusion induced increased MUC1 expression. Furthermore, MUC1 knockdown induced significant reduction of invasive and migration properties of renal cancer cells under hypoxia. Altogether, these results show that MUC1 is directly regulated by HIF-1alpha and affects the invasive and migration properties of renal cancer cells. Thus, MUC1 could serve as a potential therapeutic target in cRCC.
    Cancer Research 07/2009; 69(14):5707-15. · 8.65 Impact Factor
  • Ejc Supplements - EJC SUPPL. 01/2009; 7(2):373-373.
  • Ejc Supplements - EJC SUPPL. 01/2009; 7(2):373-373.
  • Gastroenterology 01/2009; 136(5). · 12.82 Impact Factor
  • Journal de Chirurgie. 05/2008; 145(3):289–290.
  • [Show abstract] [Hide abstract]
    ABSTRACT: To investigate whether the number of lymph nodes metastasis (LNMs) and the ratio between metastatic and examined lymph nodes (LNs) are better prognostic factors when compared with traditional staging systems in patients with esophageal carcinoma. The accuracy of the 6th UICC/TNM classification is suboptimal, especially when not taking into account neoadjuvant therapy and lymphadenectomy extent. For 536 patients who underwent curative en bloc esophagectomy, in whom 51.5% (n = 276) received neoadjuvant chemoradiation, LNMs were classified according to the 6th UICC/TNM classification and systems based on the number (< or =4 and >4) or the ratio (< or =0.2 and >0.2) of LNMs. Survival of the respective stages, predictors of survival, and influence of both chemoradiation and number of examined LNs were studied. After a median follow-up of 50 months, the 5-year survival rates were 47% for the entire population, significantly poorer for patients with >4 LNMs (8% vs. 53%, P < 0.001) or a ratio of LNMs >0.2 (22% vs. 54%, P < 0.001). After adjustment for confounding variables, a number of LNMs >4 and a ratio of LNMs >0.2 were the only predictors of poor prognosis. The prognostic role of both the number and the ratio of LNMs was maintained whether patients received neoadjuvant chemoradiation or not. Moreover, LN ratio is shown to be more accurate for inadequately staged patients (<15 examined LNs), whereas the number of LNMs is pertinent for adequately staged patients (> or =15 examined LNs). Staging systems for esophageal cancer that use the number (< or =4 or >4) and the ratio (< or =0.2 or >0.2) of LNMs have greater prognostic importance than the current staging systems because of the good stratification of the groups and their clinical utility, taking into account neoadjuvant therapy and lymphadenectomy extent.
    Annals of Surgery 03/2008; 247(2):365-71. · 6.33 Impact Factor
  • N Briez, C Mariette
    Journal de Chirurgie 01/2008; 145(2):147-52. · 0.50 Impact Factor