P A Clavien

University of Zurich, Zürich, Zurich, Switzerland

Are you P A Clavien?

Claim your profile

Publications (166)866.53 Total impact

  • [Show abstract] [Hide abstract]
    ABSTRACT: To determine MR-imaging features for the differentiation between hepatocellular carcinoma (HCC) and benign hepatocellular tumors in the non-cirrhotic liver. 107 consecutive patients without liver cirrhosis (46 male; 45±14 years) who underwent liver resection due to suspicion of HCC were included in this multi-center study. The following imaging features were assessed: lesion diameter and demarcation, satellite-lesions, central-scar, capsule, fat-content, hemorrhage, vein-infiltration and signal-intensity (SI) on native T1-, T2- and dynamic-enhanced T1-weighted images (center versus periphery). In addition, contrast-media (CM) uptake in the liver specific phase was analyzed in a sub-group of 42 patients. Significant differences between HCC (n=55) and benign lesions (n=52) were shown for native T1-, T2- and dynamic-enhanced T1-SI, fat-content, and satellite-lesions (all, P<.05). Independent predictors for HCC were T1-hypointensity (odds-ratio, 4.81), T2-hypo-/hyperintensity (5.07), lack of central tumor-enhancement (3.36), and satellite-lesions (5.78; all P<0.05). Sensitivity and specificity of HCC was 91% and 75% respectively for two out-of four independent predictors, whereas specificity reached 98% for all four predictors. Sub-analysis, showed significant differences in liver specific CM uptake between HCC (n=18) and benign lesions (n=24; P<0.001) and revealed lack of liver specific CM uptake (odds-ratio, 2.7) as additional independent feature for diagnosis of HCC. Independent MRI features indicating HCC are T1-hypointensity, T2-hypo- or hyperintensity, lack of central tumor-enhancement, presence of satellite-lesions and lack of liver specific CM-uptake. These features may have the potential to improve the diagnosis of HCC in the non-cirrhotic liver. Copyright © 2015. Published by Elsevier Ireland Ltd.
    European journal of radiology 07/2015; 84(10). DOI:10.1016/j.ejrad.2015.06.029 · 2.37 Impact Factor
  • Source
    C Tschuor · J Metzger · P -A Clavien · R Vonlanthen · K Lehmann ·
    [Show abstract] [Hide abstract]
    ABSTRACT: Inguinal hernia repair is one of the most common procedures in visceral surgery, and an important teaching operation for residents during their first years. A variety of surgical approaches is currently available, including open surgery with or without mesh and laparoscopic surgery. Here we assessed the current clinical practice for inguinal hernia surgery in Switzerland and the impact on training of surgical residents. An anonymous online survey was performed among surgical clinics of the Swiss Society of Visceral Surgery (SSVS). The overall response rate was 51 %. Nearly all hernia repairs are performed with prosthetic material, and only 3.2 % of the procedures use no mesh. Overall, open surgery is used for 58.5 % of hernias and 41.5 % are operated laparoscopically. In laparoscopic surgery, TEP is the first choice. Overall, the Lichtenstein repair is the classical teaching operation performed by residents in 77.3 % of cases. In contrast to open surgery, laparoscopic hernia repair is not a training operation and residents perform only 9.7 % of laparoscopic hernia repairs. The survey confirms the use of prosthetic material as the standard, and the Lichtenstein repair as the first choice for primary inguinal hernia repair. The popularity of laparoscopic hernia surgery is increasing at the price of less teaching operations available for young residents.
    Hernia 05/2015; 19(5). DOI:10.1007/s10029-015-1385-2 · 2.05 Impact Factor
  • O Pittet · A Nocito · H Balke · C Duvoisin · P.A. Clavien · N Demartines · D Hahnloser ·
    [Show abstract] [Hide abstract]
    ABSTRACT: According to the French Greccar III randomized trial, full mechanical bowel preparation (MBP) for rectal surgery decreases the rate of postoperative morbidity, in particular postoperative infectious complications, but MBP is not well tolerated by the patient. The aim of the present study was to determine whether a preoperative rectal enema (RE) might be an alternative to MBP. An analysis was performed of 96 matched cohort patients undergoing rectal resection with primary anastomosis and protective ileostomy at two different university teaching hospitals, whose rectal cancer management were comparable except for the choice of preoperative bowel preparation (MBP or rectal enema [RE]). Prospective databases were retrospectively analyzed. Patients were well matched for age, gender, body mass index (BMI) and Charlson's index. The surgical approach and cancer characteristics (level above anal verge, stage, and use of neoadjuvant therapy) were comparable between the two groups. Anastomotic leakage occurred in 10% of patients having MBP and in 8% having RE (p=1.00). Pelvic abscess formation (6% vs. 2%, p=0.63) and wound infection (8% vs. 15%, p=0.55) were also comparable. Extra-abdominal infection (13% vs. 13%, p=1.00) and non-infectious abdominal complications such as ileus and bleeding (27% and 31%, p=0.83) were not significantly different. Overall morbidity was comparable in the two groups (50% vs 54% [p=0.83).). A simple rectal enema before rectal surgery seems not to be associated with more postoperative infectious complications nor a higher overall morbidity than MBP. This article is protected by copyright. All rights reserved. This article is protected by copyright. All rights reserved.
    Colorectal Disease 04/2015; 17(11). DOI:10.1111/codi.12974 · 2.35 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Epidural analgesia (EDA) is a common analgesia regimen in liver resection, and is accompanied by sympathicolysis, peripheral vasodilatation and hypotension in the context of deliberate intraoperative low central venous pressure. This associated fall in mean arterial pressure may compromise renal blood pressure autoregulation and lead to acute kidney injury (AKI). This study investigated whether EDA is a risk factor for postoperative AKI after liver surgery. The incidence of AKI was investigated retrospectively in patients who underwent liver resection with or without EDA between 2002 and 2012. Univariable and multivariable analyses were performed including recognized preoperative and intraoperative predictors of posthepatectomy renal failure. A series of 1153 patients was investigated. AKI occurred in 8·2 per cent of patients and was associated with increased morbidity (71 versus 47·3 per cent; P = 0·003) and mortality (21 versus 0·3 per cent; P < 0·001) rates. The incidence of AKI was significantly higher in the EDA group (10·1 versus 3·7 per cent; P = 0·003). Although there was no significant difference in the incidence of AKI between patients undergoing minor hepatectomy with or without EDA (5·2 versus 2·7 per cent; P = 0·421), a substantial difference in AKI rates occurred in patients undergoing major hepatectomy (13·8 versus 5·0 per cent; P = 0·025). In multivariable analysis, EDA remained an independent risk factor for AKI after hepatectomy (P = 0·040). EDA may be a risk factor for postoperative AKI after major hepatectomy. © 2015 BJS Society Ltd Published by John Wiley & Sons Ltd.
    British Journal of Surgery 04/2015; 102(7). DOI:10.1002/bjs.9810 · 5.54 Impact Factor
  • M. Langiewicz · A. Schlegel · H. Bostjan · R. Graf · P.-A. Clavien ·

    Journal of Hepatology 04/2015; 62:S465. DOI:10.1016/S0168-8278(15)30617-6 · 11.34 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The body is dependent on the exogenous supply of omega-3 polyunsaturated fatty acids (n3-PUFA). These essential fatty acids are key players in regulating metabolic signaling but also exert anti-inflammatory and anti-carcinogenic properties. The liver is a major metabolic organ involved in fatty acid metabolism. Under experimental conditions, n3-PUFA exert beneficial effect on hepatic steatosis, regeneration and inflammatory insults such as ischemic injury after surgery. Some of these effects have also been observed in human subjects. However, it is unclear whether perioperative administration of n3-PUFA is sufficient to protect the liver from ischemic injury. Therefore, we designed a randomized controlled trial (RCT) assessing n3-PUFA (pre-) conditioning strategies in patients scheduled for liver surgery. The Omegaven™ trial is a multi-centric, double-blind, randomized, placebo- controlled trial applying two single doses of Omegaven™ or placebo on 258 patients undergoing major liver resection. Primary endpoints are morbidity and mortality one month after hospital discharge, defined by the Clavien- Dindo classification of surgical complications (Ann Surg 240(2):205–13, 2004) as well as the Comprehensive Complication Index (CCI) (Ann Surg 258(1):1–7, 2013). Secondary outcome variables include length of Intensive Care Unit (ICU) and hospital stay, postoperative liver function tests, fatty acid and eicosanoid concentration, inflammatory markers in serum and in liver tissue. An interim analysis is scheduled after the first 30 patients per randomization group. Long-term administration of n3-PUFA have a beneficial effect on metabolism and hepatic injury. Patients often require surgery without much delay, thus long-term n3-PUFA uptake is not possible. Also, lack of compliance may lead to incomplete n3-PUFA substitution. Hence, perioperative Omegaven™ may provide an easy and controllable way to ensure hepaative application of tic protection. Trial registration ClinicalTrial.gov: ID: NCT01884948, registered June 14, 2013; Institution Ethical Board Approval: KEK-ZH-Nr. 2010–0038; Swissmedic Notification: 2012DR3215.
    Journal of Hepatology 04/2015; 62(1):S856. DOI:10.1016/S0168-8278(15)31512-9 · 11.34 Impact Factor
  • E. Kachaylo · C. Tschuor · M. Linecker · P. Limani · M. Foti · R. Graf · B. Humar · P.-A. Clavien ·

    Journal of Hepatology 04/2015; 62:S466. DOI:10.1016/S0168-8278(15)30620-6 · 11.34 Impact Factor
  • H. Petrowsky · P.-A. Clavien ·

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: On behalf of the ALPPS Registry Group Objectives: To assess safety and outcomes of the novel 2-stage hepatectomy, Associating Liver Partition and Portal Vein Ligation for Staged Hepatectomy (ALPPS), using an international registry. Background: ALPPS induces accelerated growth of small future liver rem-nants (FLR) to allow curative resection of liver tumors. There is concern about safety based on reports of higher morbidity and mortality. Methods: A Web-based data entry system was created with password access and data pseudoencryption (NCT01924741). All patients with com-plete 90-day data were included. Multivariate logistic regression analysis was performed to identify independent risk factors for severe complications and mortality and volume growth of the FLR. Results: Complete data were available for 202 patients. A total of 141 (70%) patients had colorectal liver metastases (CRLM). Median starting standard-ized future liver remnants of 21% increased by 80% within a median of 7 days. Ninety-day mortality was 19/202 (9%). Severe complications including mortalities (Clavien-Dindo ≥IIIb) occurred in 27% of patients. Independent factors for severe complications were red blood cell transfusion [odds ratio (OR), 5.2), ALLPS stage I operating time greater than 300 minutes (OR, 4.4), age more than 60 years (OR, 3.8), and non-CRLM (OR, 2.7). Age, use of Pringle maneuver, and histologic changes led to less volume growth. In patients younger than 60 years with CRLM, 90-day mortality was similar to conventional 2-stage hepatectomies for CRLM. Conclusions: This is the first analysis of the ALPPS registry showing that ALPPS shows increased perioperative morbidity and mortality in older patients but better outcomes in patients with CRLM.
    Annals of surgery 11/2014; 260(5). DOI:10.1097/SLA.0000000000000947 · 8.33 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Aim-Background Liver transplantation is an acceptable therapy in selective patients with hepatic epithelioid haemangioendothelioma (HEH) and has been reported in absent or limited extrahepatic disease. The purpose of this paper is to present an interesting case of a new surgical technique involving the use of an inferior vena cava graft from a 24-hour deceased donor in a patient who underwent liver transplantation from a living donor. Case report-Methods We describe a 45-year-old woman with rapidly growing HEH and tumour involvement of all hepatic veins with obstruction of the inferior cava. The patient underwent living donor liver transplantation (LDLT). In view of the need for urgent liver replacement and the lack of banked vessels, we used a cava graft obtained from a blood-group matched man, who died twenty-four hours prior to LDLT. His cadaver was temporarily cold and stored in the Department of Pathology prior to his funeral. The recipient was put on mTOR inhibitor and carefully monitored. Results The cava interpolate remained uneventfully open. Six months after surgery, the patient developed an asymptomatic partial thrombosis of 30% at interpolation which was treated with low molecular weight heparin. Eighteen months after the initial operation, the patient died due to diffuse disease recurrence. Conclusions Following a review of the literature, this case emerges as the first description (new technique) of the use of a venous graft derived from a stored human cadaver, twenty-four hours after cardiac arrest. Questions arise as to whether the solution to technical surgical problems is the only safe and effective parameter in achieving the best treatment modality for similar cases.
    Hellēnikē cheirourgikē. Acta chirurgica Hellenica 11/2014; 85(6):431-436. DOI:10.1007/s13126-013-0081-x
  • Article: Reply
    P. Starlinger · M. Lesurtel · C. Brostjan · P. A. Clavien · T. Gruenberger ·

    Hepatology 10/2014; DOI:10.1002/hep.27574 · 11.06 Impact Factor
  • Source

    Annals of Surgery 08/2014; 260(2):e4. DOI:10.1097/SLA.0000000000000735 · 8.33 Impact Factor
  • Source
    P. Dutkowski · P.‐A. Clavien ·

    British Journal of Surgery 06/2014; 101(7). DOI:10.1002/bjs.9540 · 5.54 Impact Factor
  • Source
    P.‐A. Clavien · Milo A Puhan ·

    British Journal of Surgery 05/2014; 101(6). DOI:10.1002/bjs.9499 · 5.54 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: We describe a case of a 62-year-old diabetic woman with hepatocellular carcinoma due to chronic hepatitis B virus infection. Two weeks after orthotopic liver transplantation, endoscopy for massive upper gastrointestinal bleeding revealed a large necrotic area in the gastric fundus. The patient underwent emergency resection. Histopathologically, angioinvasive mold infection compatible with mucormycosis was diagnosed in a large area of necrosis, mimicking an atypically localized gastric ulcer. Foreign bodies originating from transarterial chemoembolization (TACE) performed 7 and 8 months earlier and 40 days before transplantation were identified in the submucosal tissue. The patient was treated with liposomal amphotericin B (LAB) for 5 weeks, followed by 7 weeks of posaconazole. Follow-up biopsies after 1 and 5 months confirmed successful treatment. Review of the radiological images of the TACE procedure showed that some of the TACE material had been diverted to the stomach via an accessory gastric branch originating from the left hepatic artery. TACE agents may be associated with chronic, refractory gastroduodenal ulcers. We hypothesize that the ischemic lesion was first colonized with presumed Mucorales mold and invasive growth was promoted by the posttransplantation immunosuppression. Careful exploration of extrahepatic collaterals during TACE may prevent this complication.
    Infection 03/2014; 42(4). DOI:10.1007/s15010-014-0603-7 · 2.62 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: REFINE was a 12-month, prospective, open-label study in 356 patients receiving de novo liver transplantation for hepatitis C virus (HCV) cirrhosis, randomized to cyclosporine A (CsA) or tacrolimus with (i) no steroids, IL-2 receptor antibody induction and mycophenolic acid, or (ii) slow steroid tapering. The primary analysis population based on availability of liver biopsies comprised 165 patients (88 CsA, 77 tacrolimus). There was no difference in the primary endpoint, fibrosis stage ≥2 at 12 months, which occurred in 63/88 CsA-treated patients (71.6%) and 52/77 tacrolimus-treated patients (67.5%) (odds ratio [OR] 1.11; 95% CI 0.56, 2.21; p = 0.759). Similarly, no significant between-group difference occurred at month 24 (OR 1.15; 95% CI 0.47, 2.80; p = 0.767). Among steroid-free patients, fibrosis score ≥2 was significantly less frequent with CsA versus tacrolimus at month 12 (7/37 [18.9%] vs. 16/38 [42.1%]; p = 0.029). HCV viral load was similar in both the tacrolimus- and CsA-treated cohorts. Mean blood glucose was significantly higher with tacrolimus from day 15 onward. Biopsy-proven acute rejection, graft loss and death were similar. These results showed no differences in posttransplant HCV-induced liver fibrosis between patients treated with CsA or tacrolimus in steroid-containing regimens, whereas CsA in steroid-free protocols was associated with reduced severity of fibrosis progression at 1 year posttransplant.
    American Journal of Transplantation 03/2014; 14(3). DOI:10.1111/ajt.12620 · 5.68 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Intraoperative adverse events significantly influence morbidity and mortality of laparoscopic colorectal resections. Over an 11-year period, the changes of occurrence of such intraoperative adverse events were assessed in this study. Analysis of 3,928 patients undergoing elective laparoscopic colorectal resection based on the prospective database of the Swiss Association of Laparoscopic and Thoracoscopic Surgery was performed. Overall, 377 intraoperative adverse events occurred in 329 patients (overall incidence of 8.4 %). Of 377 events, 163 (43 %) were surgical complications and 214 (57 %) were nonsurgical adverse events. Surgical complications were iatrogenic injury to solid organs (n = 63; incidence of 1.6 %), bleeding (n = 62; 1.6 %), lesion by puncture (n = 25; 0.6 %), and intraoperative anastomotic leakage (n = 13; 0.3 %). Of note, 11 % of intraoperative organ/puncture lesions requiring re-intervention were missed intraoperatively. Nonsurgical adverse events were problems with equipment (n = 127; 3.2 %), anesthetic problems (n = 30; 0.8 %), and various (n = 57; 1.5 %). Over time, the rate of intraoperative adverse events decreased, but not significantly. Bleeding complications significantly decreased (p = 0.015), and equipment problems increased (p = 0.036). However, the rate of adverse events requiring conversion significantly decreased with time (p < 0.001). Patients with an intraoperative adverse event had a significantly higher rate of postoperative local and general morbidity (41.2 and 32.9 % vs. 18.0 and 17.2 %, p < 0.001 and p < 0.001, respectively). Intraoperative surgical complications and adverse events in laparoscopic colorectal resections did not change significantly over time and are associated with an increased postoperative morbidity.
    Langenbeck s Archives of Surgery 01/2014; 399(3). DOI:10.1007/s00423-013-1156-4 · 2.19 Impact Factor
  • Article: Pancreas
    A. Wirsching · M. Lesurtel · P.-A. Clavien ·
    [Show abstract] [Hide abstract]
    ABSTRACT: About 80 % of newly diagnosed patients with pancreatic adenocarcinoma cannot benefit from a curative strategy. Palliative approaches of unresectable pancreatic cancer should be adjusted to the expected survival with the aim of preserving the quality of life of these patients. When the diagnosis of unresectable disease is made, nonsurgical endoscopic approaches should be prioritized in order to keep hospital stay as short as possible without delaying systemic chemotherapy. If an unresectable disease is diagnosed at laparotomy, an appropriate palliative surgical treatment should be considered to prevent biliary and enteral obstruction, as well as pain exacerbation due to tumour invasion. Surgical bypass procedures allow significantly more lasting palliation than endoscopic procedures in distinct situations. Since morbidity and mortality of pancreatoduodenectomy have significantly decreased in the last decades, a more aggressive approach towards palliative resection could be justified in specific circumstances. Pain control should not be neglected and is optimized when pharmacotherapy and chemical neurolysis are associated. Since palliative treatment of unresectable pancreatic cancer is not trivial, the choice of the best approach should be discussed by a multidisciplinary team including surgeons, gastroenterologists, radiologists, oncologists and physicians in charge of palliative care.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Virtual reality (VR) simulators are widely used to familiarize surgical novices with laparoscopy, but VR training methods differ in efficacy. In the present trial, self-controlled basic VR training (SC-training) was tested against training based on peer-group-derived benchmarks (PGD-training). First, novice laparoscopic residents were randomized into a SC group (n = 34), and a group using PGD-benchmarks (n = 34) for basic laparoscopic training. After completing basic training, both groups performed 60 VR laparoscopic cholecystectomies for performance analysis. Primary endpoints were simulator metrics; secondary endpoints were program adherence, trainee motivation, and training efficacy. Altogether, 66 residents completed basic training, and 3,837 of 3,960 (96.8 %) cholecystectomies were available for analysis. Course adherence was good, with only two dropouts, both in the SC-group. The PGD-group spent more time and repetitions in basic training until the benchmarks were reached and subsequently showed better performance in the readout cholecystectomies: Median time (gallbladder extraction) showed significant differences of 520 s (IQR 354-738 s) in SC-training versus 390 s (IQR 278-536 s) in the PGD-group (p < 0.001) and 215 s (IQR 175-276 s) in experts, respectively. Path length of the right instrument also showed significant differences, again with the PGD-training group being more efficient. Basic VR laparoscopic training based on PGD benchmarks with external assessment is superior to SC training, resulting in higher trainee motivation and better performance in simulated laparoscopic cholecystectomies. We recommend such a basic course based on PGD benchmarks before advancing to more elaborate VR training.
    World Journal of Surgery 08/2013; 37(11). DOI:10.1007/s00268-013-2175-6 · 2.64 Impact Factor
  • C Tschuor · P Limani · A Nocito · D Dindo · P -A Clavien · D Hahnloser ·
    [Show abstract] [Hide abstract]
    ABSTRACT: Background: Perineal stapled prolapse (PSP) resection is a novel operation for treating external rectal prolapse. However, no long-term results have been reported in the literature. This study analyses the long-term recurrence rate, functional outcome, and morbidity associated with PSP resection. Methods: Nine consecutive patients undergoing PSP resection between 2007 and 2011 were prospectively followed. Surgery was performed by the same surgeons in a standardised technique. Recurrence rate, functional outcome, and complication grade were prospectively assessed. Results: All 9 patients undergoing PSP resection were investigated. The median age was 72 years (range 25-88 years). No intraoperative complications occurred. Faecal incontinence, preoperatively present in 2 patients, worsened postoperatively in one patient (Vaizey 18-22). One patient developed new-onset faecal incontinence (Vaizey 18). The median obstructive defecation syndrome score decreased postoperatively significantly from 11 (median; range 8-13) to 5 (median; range 4-8) (p < 0.005). At a median follow-up of 40 months (range 14-58 months), the prolapse recurrence rate was 44 % (4/9 patients). Conclusions: The PSP resection is a fast and safe procedure associated with low morbidity. However, the poor long-term functional outcome and the recurrence rate of 44 % warrant a cautious patient selection.
    Techniques in Coloproctology 04/2013; 17(5). DOI:10.1007/s10151-013-1009-8 · 2.04 Impact Factor

Publication Stats

7k Citations
866.53 Total Impact Points


  • 2001-2015
    • University of Zurich
      • • Center for Microscopy and Image Analysis - ZMB
      • • Internal Medicine Unit
      • • Klinik für Viszeral- und Transplantationschirurgie
      Zürich, Zurich, Switzerland
  • 2014
    • Université Paris-Sud 11
      Orsay, Île-de-France, France
  • 2001-2013
    • University Hospital Zürich
      Zürich, Zurich, Switzerland
  • 1996-2002
    • Duke University Medical Center
      • • Department of Surgery
      • • Department of Pathology
      Durham, North Carolina, United States
  • 1996-2000
    • Duke University
      • • Department of Surgery
      • • Department of Medicine
      Durham, North Carolina, United States
  • 1989-1996
    • University of Toronto
      • Department of Surgery
      Toronto, Ontario, Canada
  • 1993
    • Washington University in St. Louis
      • Department of Surgery
      San Luis, Missouri, United States
  • 1991-1992
    • Samuel Lunenfeld Research Institute
      Toronto, Ontario, Canada
  • 1988-1989
    • University of Geneva
      • Department of Surgery
      Genève, Geneva, Switzerland