Search

Background: Postoperative mortality commonly is defined as death occurring within 30 days of surgery or during hospitalization. After resection for liver malignancies, this definition may result in underreporting, because mortality caused by postoperative complications can be delayed as the result of improved critical care. The aim of this study was to estimate statistically the acute postoperative period (APP) after partial hepatectomy and to compare mortality within this phase to standard timestamps., Methods: From a prospective database, 784 patients undergoing resection for primary and secondary hepatic malignancies between 2003 and 2013 were reviewed. For estimation of APP, a novel statistical method applying tests for a constant postoperative hazard was implemented. Multivariable mortality analysis was performed., Results: The APP was determined to last for 80 postoperative days (95% confidence interval 40-100 days). Within this period, 55 patients died (7.0%; 80-day mortality). In comparison, 30-day mortality (N = 32, 4.0%) and in-hospital death (N = 39, 5.0%) were relevantly less. No patient died between postoperative days 80 and 90. The causes of mortality within 30 days and from days 30-80 did not greatly differ, especially regarding posthepatectomy liver failure (44% vs 39%, P = .787). Septic complications, however, tended to cause late deaths more frequently (43% vs 25%, P = .255). Comorbidities (Charlson comorbidity index ≥ 3; P = .046), increased preoperative alanine aminotransferase activity (P = .030), and major liver resection (P = .035) were independent risk factors of 80-day mortality., Conclusion: After liver resection for primary and secondary malignancies, 90-day rather than 30-day or in-hospital mortality should be used to avoid underreporting of deaths., (Copyright © 2015 Elsevier Inc. All rights reserved.)

Aims: To determine predictors of failed enhanced recovery after surgery (ERAS) in patients after elective colorectal surgery., Methods: A cohort of 55 patients undergoing elective colorectal surgery was monitored prospectively. Perioperative care was based on a previously established protocol for ERAS. Pre-, intra-, and postoperative parameters were analyzed to elicit predictors of ERAS failure. ERAS failure was defined as prolonged hospital stay (> 7 days). The risk calculator CR-POSSUM was evaluated for its clinical utility., Results: Body mass index (BMI) or the American Society of Anesthesiologists score (ASA) was not associated with ERAS failure on univariate analysis, but patients that failed ERAS were significantly older (64 y vs 54 y; p = 0.023). Prolonged length of stay (>7 days) was also associated with an open approach (p = 0.009), intraoperative nasogastric tube placement (p = 0.005), blood loss > 500 ml (p = 0.008), stoma formation (p = 0.006) and insertion of more than one intraabdominal drain during surgery (p = 0.005). Postoperative continuation of intravenous fluids (p = 0.027), reinsertion of urinary catheter (p = 0.045) and postoperative ileus (p = 0.020) were also strongly associated with delayed discharge on univariate analysis. After multivariate analysis the preoperative parameters CR-POSSUM score (p = 0.022), increasing BMI (p = 0.014) and preoperative albumin level (p = 0.031) were all independently associated with failure of ERAS., Conclusions: A variety of perioperative factors contribute to failure of ERAS in routine practice. CR-POSSUM can help to identify patients at risk for possible failure of ERAS. This may help to optimize avoidable factors, or accommodate those patients likely to require a longer post-operative stay.

Purpose: The purpose of this study was to assess the impact of surgical training and surgeon's experience on surgical quality and early outcome parameters., Methods: The outcomes of 184 deceased donor kidney transplant procedures were retrospectively reviewed. Intraoperative quality parameters, complication rates, and early outcome parameters were analyzed with respect to surgeon's experience. Surgeons were grouped according to their level of experience: trainees (n = 5), who were senior residents in their transplant rotation; low-experience surgeons (n = 4), who had an individual caseload of at least 30 supervised procedures; medium-experience surgeons (n = 5), who had an individual caseload of a minimum of 30 unsupervised (+30 supervised) procedures; and high-experience surgeons (n = 2), who had an individual caseload of more than 200 procedures., Results: Surgical training was offered in 20 % of all cases. Surgeons with a high experience in kidney transplantation had a significantly lower warm ischemic time (30 versus 35 min, p = 0.01) and total operation time (135 versus 155 min, p = 0.43) as compared to the other study groups. However, this did not translate into a better outcome after transplantation. The complication rate and early outcome parameters such as length of hospital stay, serum creatinine at discharge, and graft and patient survival at 3 months post TX did not show any significant difference between the groups., Conclusion: We conclude that hands-on training of residents in kidney transplantation is feasible and surgical experience can be safely gained within a staged surgical training program.

Background: Cytomegalovirus (CMV) is a common opportunistic infection after solid organ transplantation. Cytomegalovirus causes increased morbidity, mortality, and reduced allograft survival. Prophylaxis may help control the virus but is associated with substantial side effects and does not completely prevent virus reactivation; relapses after cessation of the prophylaxis are frequent. Experimental and clinical data suggest that mTOR inhibitors may have an anti-CMV effect. Here, we present a meta-analysis of clinical trials after solid organ transplantation and describe potential mechanisms involved in the anti-CMV effect of mTOR-inhibitors., Methods: The current literature was reviewed for randomized controlled trials in solid organ transplantation comparing an mTOR-I with a non-mTOR-I (CNI based) treatment. The scientific quality of the trials was assessed by the Jadad score, the use of an effective allocation concealment (AC) and the existence of an intention-to-treat (ITT) analysis. Cytomegalovirus incidence was assessed in studies comparing 1) an mTOR-I-based with a CNI-based immunosuppression (10 trials, n=3,100 patients) and 2) an mTOR-I/CNI combination therapy with a CNI-based immunosuppression (15 trials, n=7,100 patients)., Results: In the first meta-analysis, CMV events after solid organ transplantation occurred significantly more often under CNIs (RR=2.27). The second meta-analysis comparing the mTOR-I + CNI combination with a CNI treatment in 15 trials of kidney, heart, and liver transplantation showed again a higher CMV incidence when patients received an mTOR-I free immunosuppression (RR=2.45)., Conclusions: mTOR-inhibitor treatment either alone or in combination with CNIs reduces significantly the CMV incidence after organ transplantation. With the use of an mTOR-inhibitor, CMV prophylaxis may be dispensible.

Background: In patients with Crohn's disease (CD), the effect of anti-tumor necrosis factor alpha (TNF-α) antibody therapy on postoperative complications remains unclear. We aimed to determine the effects of infliximab on postoperative complication rates in patients undergoing abdominal surgery for CD., Methods: Infliximab-treated CD patients undergoing abdominal surgery were identified in a prospective database. Gender- and age-matched CD patients without infliximab treatment served as controls. General and complication-related information was retrieved from patient records., Results: Forty-eight patients underwent abdominal surgery within 3 months (median 60 days, range 1-90 days) after infliximab administration (56% female, median age 35 years, range 17-66 years). Forty-eight patients without infliximab served as controls (50% female, 39 [17-68] years). Patient characteristics and number of minor complications were comparable between groups: wound infection (infliximab: 19% vs. controls: 15%), prolonged postoperative ileus (15% vs. 4%), and urinary tract infection (2% vs. 0%; all P > 0.05). No differences were found in major complications: anastomotic leakage (infliximab: 4% vs. controls: 13%), abscess formation (6% vs. 10%), bowel perforation (2% vs. 4%), stoma complication (6% vs. 2%), postoperative hemorrhage (8% vs. 2%), and enterocutaneous fistula (4% vs. 0%; all P > 0.05). One malnourished infliximab-treated patient with a complicated course of disease died postoperatively after anastomotic leakage, sepsis, and cardiac arrhythmia. Eleven infliximab and 10 control patients required reoperation (P > 0.05). Hospital stay was comparable between groups (infliximab: 13 [5-41] vs. controls: 12 [5-54] days; P > 0.05)., Conclusions: Infliximab does not affect postoperative complication rates, suggesting no need to alter surgical management in these patients., (Copyright © 2011 Crohn's & Colitis Foundation of America, Inc.)

Background: Increased risks related to surgery might reflect the nutritional status of some patients. Such a group might benefit from perioperative nutritional support. The purpose of this study was to identify the relative importance of nutritional risk screening along with established medical, anaesthetic and surgical predictors of postoperative morbidity and mortality., Methods: This prospective observational study enrolled consecutive eligible patients scheduled for elective abdominal operations. Data were collected on nutritional variables (body mass index, weight loss, food intake), age, sex, type and extent of operation, underlying disease, American Society of Anesthesiologists grade and co-morbidities. A modified composite nutritional screening tool (Nutritional Risk Screening, NRS 2002) currently recommended by European guidelines was used. Relative complication rates were calculated with multiple logistic regression and cumulative proportional odds models., Results: Some 653 patients were enrolled of whom 132 (20.2 per cent) sustained one or more postoperative complications. The frequency of this event increased significantly with a lower food intake before hospital admission. No other individual or composite nutritional variable provided comparable or better risk prediction (including NRS 2002). Other factors significantly associated with severe postoperative complications were ASA grade, male sex, underlying disease, extent of surgical procedure and volume of transfused red cell concentrates., Conclusion: In abdominal surgery, preoperative investigation of feeding habits may be sufficient to identify patients at increased risk of complications. Nutritional risk alone, however, is not sufficient to predict individual risk of complications reliably., (Copyright © 2012 British Journal of Surgery Society Ltd. Published by John Wiley & Sons, Ltd.)

Purpose: Bladder drainage (BD) of pancreatic transplants is associated with a unique set of complications. We intended to analyze the incidence, indications, complications and long-term results of enteric conversion procedures (EC)., Methods: Using a prospective database, 32 EC patients out of 433 simultaneous pancreas-kidney-transplant (SPK) recipients were identified. Graft and patient survival rates were compared with those after primary enteric drainage (ED)., Results: The mean SPK-EC interval was 5.0 yr, and the mean patient follow-up was 13.8 yr. Indications for EC were genitourinary symptoms (62.5%), duodenal complications (15.6%), graft pancreatitis (12.5%), pyelonephritis (6.3%), and metabolic acidosis (3.1%). All patients reported significant long-term resolution of symptoms. Surgical complications, reoperations, early graft loss, and 30-d mortality occurred in 31.3%, 25.0%, 6.3%, and 3.1% of cases, respectively. Pancreatic graft and patient survival rates at 1, 5, and 10 yr after SPK were comparable between EC patients and ED patients at the same institution., Conclusion: For the treatment of symptoms associated with BD, EC results in excellent long-term graft function and significant resolution of symptoms even years after SPK. Postoperative morbidity after EC including early reoperation and graft loss, however, has to be considered., (© 2010 John Wiley & Sons A/S.)

Background: Insulin regulates albumin synthesis in vitro and in various experimental models. The current study was undertaken to determine the effects of a physiologic hyperinsulinemia on albumin synthesis in postoperative patients in whom plasma albumin concentrations are decreased., Methods: Studies were performed in postabsorptive patients after major abdominal operations. Mass spectrometry techniques were used to directly determine the incorporation rate of 1-[(13)C]-leucine into albumin. Consecutive blood samples were taken during a continuous isotope (D-Glc) infusion (0.16 µmol/kg/min). Isotopic enrichments were determined at baseline (period I) and after a 4-hour D-glucose (D-Glc) infusion at currently recommended rates (170 mg/kg/h, n = 10) or after infusion of saline (control group, n = 8) (period II)., Results: After D-Glc infusion, plasma insulin concentrations increased significantly (period I, 6.6 ± 1.8 µU/mL; period II, 21.4 ± 2.1 µU/mL; P < .01). In contrast, plasma insulin concentration remained constant in control patients (period I, 3.8 ± 0.9 µU/mL(-1); period II, 5.9 ± 1.1 µU/mL; not significant vs period I, but P < .005 vs the corresponding value at the end of period II in the control group). Hyperinsulinemia was without effect on fractional albumin synthesis (period I, 12.8% ± 1.9%/d; period II, 11.9% ± 1.9%/d; not significant), and synthesis rates corresponded to those measured in controls (period I, 13.0% ± 1.2%/d; period II, 12.1% ± 0.1%/d; not significant vs period I and vs D-Glc infusion)., Conclusions: A standard D-Glc infusion is insufficient to increase albumin synthesis in postoperative patients.

Creation of a protective stoma is nowadays considered the standard of care in patients undergoing low rectal resection to protect these patients from the potentially hazardous consequences of an anastomotic leak. This appears reasonable in patients with acknowledged risk factors such as male gender, low anastomosis, preoperative radiochemotherapy, intraoperative complications, or steroid treatment to ensure patients' safety. However, from our view, it is debatable, if patients without these risk factors can undergo low rectal resection without a stoma. This approach can prevent patients form potential risks of stoma creation as well as closure and the associated readmission to the hospital. Based on reliable patient selection, avoiding a protective stoma during low rectal resection can increase patients' satisfaction and decrease primary and secondary medical costs. However, this approach is hampered by the lack of evidence for patient selection, leading to legal concerns that justify this approach only in highly motivated patients after detailed counseling of the individual patient.

Background: Pancreatic fistula (PF) represents a major complication after distal pancreatectomy. In a consecutive series of 110 patients, risk factors for the incidence of PF and surgical morbidity were identified., Methods: Patients having undergone distal pancreatectomy between 2003 and 2007 were identified. Clinicopathologic parameters as well as perioperative data were correlated with the incidence of PF and overall surgical morbidity using univariate and multivariate models., Results: In 72 patients (65%), malignant disease was present. Splenectomy and multivisceral resection were performed in 84 (76%) and 47 (42%) patients, respectively. Overall major surgical morbidity was 18%, and 12 patients (11%) developed PFs. A body mass index > 25 kg/m(2) was the only independent significant predictive factor for PF. Malignancy, splenectomy, multivisceral resection, transfusion, comorbidity, and stapler use did not show statistical significance. For overall surgical morbidity, there was no significant indicator., Conclusions: A body mass index > 25 kg/m(2) contributes to the incidence of PF after distal pancreatectomy. Other parameters did not show a significant influence on PF or on overall surgical morbidity., (2010 Elsevier Inc. All rights reserved.)

Background: Postoperative adhesions occur when fibrous strands of internal scar tissue bind anatomical structures to one another. The most common cause of intra-abdominal adhesions is previous intra-abdominal surgical intervention. Up to 74% of intestinal obstructions are caused by post surgical adhesions. Although a variety of methods and agents have been investigated to prevent post surgical adhesions, the problem of peritoneal adhesions remains largely unsolved. Materials serving as an adhesion barrier are much needed., Methods/design: This is a prospective, randomised, controlled, patient blinded and observer blinded, single centre phase I-II trial, which evaluates the safety of A-Part Gel as an adhesion prophylaxis after major abdominal wall surgery, in comparison to an untreated control group. 60 patients undergoing an elective median laparotomy without prior abdominal surgery are randomly allocated into two groups of a 1:1- ratio. Safety parameter and primary endpoint of the study is the occurrence of wound healing impairment or peritonitis within 28 (+10) days after surgery. The frequency of anastomotic leakage within 28 days after operation, occurrence of adverse and serious adverse events during hospital stay up to 3 months and the rate of adhesions along the scar within 3 months are defined as secondary endpoints. After hospital discharge the investigator will examine the enrolled patients at 28 (+10) days and 3 months (+/-14 days) after surgery., Discussion: This trial aims to assess, whether the intra-peritoneal application of A-Part Gel is safe and efficacious in the prevention of post-surgical adhesions after median laparotomy, in comparison to untreated controls.

Background: The Surgical Department of the University Hospital Grosshadern has been making a systematic record of complications since 2005. With respect to the ongoing problem of under-financing from DRG reimbursements, an analysis of the relationship between surgical cases with severe complications and insufficient reimbursement warranted a detailed analysis., Material and Methods: Out of 16 762 in-house patients during 2005-2007 we assigned 6707 cases into four divisions - hepato-pancreato-biliary, colorectal, minimal invasive and general abdominal surgery as well as 1469 cases of thoracic surgery, for costs and reimbursement. In all groups patients with mandatory treament of complications were compared to the remaining cases without complications. Within these, further subgroups were analysed: patients with a need for artificial ventilation (partition A of the G-DRG system), cases with excessive loss (underfunding above 10 000 Euro) and their intersections., Results: With the exception of minimal invasive surgery, each division featured 10-15 % of serious complications. Losses for these cases ranged from 159 % (thoracic surgery) to 102 % (other abdominal surgery) of the overall loss in each division. Cases with excessive losses, representing 1.5 % of all patients, caused 80 % to 100 % of this deficit. Complicated cases alloted to DRGs for artificial ventilation still represented 50 % of the under-fund-ing., Conclusion: Cases with mandatory complication treatment can be discerned as separate economic entities. They are considerably overlapping cases with excessive underfunding, so further analysis might lead to an improved reimbursement policy. In addition, the connection between quality management and economic efficiency is highlighted.

In surgery, indications for artificial nutrition comprise prevention and treatment of catabolism and malnutrition. Thus in general, food intake should not be interrupted postoperatively and the re-establishing of oral (e.g. after anastomosis of the colon and rectum, kidney transplantation) or enteral food intake (e.g. after an anastomosis in the upper gastrointestinal tract, liver transplantation) is recommended within 24 h post surgery. To avoid increased mortality an indication for an immediate postoperatively artificial nutrition (enteral or parenteral nutrition (PN)) also exists in patients with no signs of malnutrition, but who will not receive oral food intake for more than 7 days perioperatively or whose oral food intake does not meet their needs (e.g. less than 60-80%) for more than 14 days. In cases of absolute contraindication for enteral nutrition, there is an indication for total PN (TPN) such as in chronic intestinal obstruction with a relevant passage obstruction e.g. a peritoneal carcinoma. If energy and nutrient requirements cannot be met by oral and enteral intake alone, a combination of enteral and parenteral nutrition is indicated. Delaying surgery for a systematic nutrition therapy (enteral and parenteral) is only indicated if severe malnutrition is present. Preoperative nutrition therapy should preferably be conducted prior to hospital admission to lower the risk of nosocomial infections. The recommendations of early postoperative re-establishing oral feeding, generally apply also to paediatric patients. Standardised operative procedures should be established in order to guarantee an effective nutrition therapy.

Background: Leakage from the pancreaticojejunostomy is the major cause of septic complications after partial pancreaticoduodenectomy. This study evaluated a new transpancreatic U-suture technique (Blumgart anastomosis, BA), which aims to avoid shear forces during knot-tying., Methods: Using a before-after study design, BA was compared with a modified Cattell-Warren anastomosis (CWA). Two patient cohorts (CWA, 90; BA, 92), which were similar with respect to primary diagnosis, age, sex and American Society of Anesthesiologists score, were compared retrospectively. Dependent variables were surgical and overall morbidity and mortality after partial pancreaticoduodenectomy., Results: Duration of operation (354 versus 328 min for CWA versus BA; P = 0.002), pancreatic leakage rate (13 versus 4 per cent; P = 0.032), postoperative haemorrhage (11 versus 3 per cent; P = 0.040), total surgical complications (31 versus 15 per cent; P = 0.011), general complications (36 versus 17 per cent; P = 0.005) and length of intensive care unit stay (median 5.4 versus 2.8 days; P = 0.015) were significantly reduced after BA. These effects were not related merely to an improvement over time., Conclusion: BA appears to be a fast, simple and safe technique for pancreaticojejunostomy. It might reduce leakage rates and surgical complications after partial pancreaticoduodenectomy., (Copyright 2009 British Journal of Surgery Society Ltd.)

The transplantation of marginal organs or those meeting the so-called extended donor criteria (EDC) is today a significant option to alleviate the low availability or organs and to increase the number of transplantation which in turn is -accompanied by a lower mortality among wait-ing-list patients. However such an extension of the spender pool carries the risks of an increased incidence of organ dysfuntions and a higher recipient mortality. This situation presents an ethical problem when marginal organs are accepted for transplantation because the anticipated mortality for the individual recipient cannot be determined. The transplantation of marginal organs from -donors with a high MELD score seems to be linked to a higher mortality. In particular, the combina-tions of high donor age and long ischaemic time or advanced donor age and hepatitis C infection in the recipient are definitively associated with a significantly poorer organ survival rate. In view of the serious lack of organs, efforts should be made, for example, by shortening of the is-chae-mic time and the development of therapeutic strategies, to improve the function and increase the number of usable marginal organs and thus to increase pool of donor organs. The refusal of marginal organs on the basis of individual EDC without consideration of the status of recipient does not seem to be adequate.

Background: Acute mortality of unselected critically ill patients has improved during the last 15 years. Whether these benefits also affect survival of critically ill patients with secondary peritonitis is unclear as is the relevance of specific prognostic factors, such as source control., Methods: We performed a retrospective analysis of data collected prospectively from March 1993 to February 2005. A cohort of 319 consecutive postoperative patients with secondary peritonitis requiring intensive care was evaluated. End points for outcome analysis were derived from daily changes of hazard rate., Results: Four-month survival rate after intensive care unit (ICU) admission was 31.7%. For patients who have survived for more than 4 months, the 1-year survival was 82.7%. After adjustment for relevant covariates, a high disease severity at ICU admission and during ICU stay, specific comorbidities (extended malignancies, liver cirrhosis) and sources of infection (distal esophagus, stomach), and an inadequate initial antibiotic therapy were associated with worse 4-month prognosis. Inability to obtain source control was the most important determinant of mortality, and treatment after 2002 was combined with improved prognosis., Conclusions: Four-month prognosis of critically ill, surgical patients with secondary peritonitis is poor and mostly determined by the ability to obtain source control. Outcome has improved since 2002, and after successful surgical and intensive care therapy long-term survival seems to be good.

Abnormalities in cardiocirculatory, respiratory, or coagulatory parameters are frequent after major surgery, but so far, no study has investigated their predictive value for early intensive care unit (ICU) mortality. We aimed to describe and quantify the relation between these parameters that are routinely determined on ICU admission and early death after complex surgery. Individual patient data were available from a local ICU database. We performed a retrospective observational cohort study using prospectively collected data from March 1, 1993, through February 28, 2005. A cohort of 4,214 cases who were admitted to the ICU immediately after operation was analyzed. We studied age, sex, number of red blood cell units transfused on admission day, and admission values for heart rate, systolic blood pressure, hemoglobin concentration, partial thromboplastin time, prothrombin time, respiratory function (Pao2/Fio2 ratio), and body temperature for their association with 4-day mortality. Effects were adjusted for the underlying disease and for disease severity during the first 24 h after admission. We used generalized additive models to fit continuous variables individually before combining them into the final generalized model. We found an independent linear association between the number of transfused red blood cell units, partial thromboplastin time, and body temperature with acute outcome. A smoothed model described the independent interaction between admission blood pressure and early death. Only values of less than 80 mmHg were associated with an increased risk of 4-day mortality. According to these results, bleeding complications after ICU admission should be treated aggressively to prevent early death of the patient. However, normotensive conditions do not seem to be required to prevent early mortality. Whether rapid rewarming may improve outcome needs further rigorous study.

Background and Aims: Significant progress in surgical technique and perioperative management has substantially reduced the mortality rate of pancreatic surgery. However, morbidity remains considerably high, even in expert hands and leakage from the pancreatic stump still accounts for the majority of surgical complications after pancreatic head resection. For that reason, management of the pancreatic remnant after partial pancreatoduodenectomy remains a challenge. This review will focus on technique, pitfalls, and complication management of pancreaticoenteric anastomoses., Materials and Methods: A medline search for surgical guidelines, prospective randomized controlled trials, systematic metaanalysis, and clinical reports was performed with regard to surgical technique and complication management of pancreatic anastomoses., Results: Pancreaticojejunostomy appears to be most widely performed, but pancreaticogastrostomy is a reasonable alternative. Postoperative treatment with octreotide can be recommended only for patients with soft pancreatic tissue, and neither stents of the pancreatic duct nor drainages have proven to effectively reduce anastomotic complications. Gastroparesis remains the most common complication after pancreatic surgery and should be treated conservatively. However, it may be a symptom of other local complications, such as anastomotic leakage, pancreatic fistula or abscess. All septic complications may finally result in late postoperative hemorrhage, which requires immediate diagnostic workup and therapy. Today, interventional radiology has emerged as a standard tool in the management of local septic complications and bleeding. Therefore, relaparotomy has become less frequent and salvage pancreatectomy is now a rare procedure in case of local complications., Conclusion: The surgeon's experience with one or the other technique of pancreatic anastomosis appears to be more important than the technique itself.

Introduction: The negative influence of conversion from laparoscopic to open colorectal resection on early postoperative morbidity and outcome has been demonstrated several times. In this study, we analyzed the conversion rate and its influence on early postoperative morbidity and short-term oncological outcome following laparoscopic rectal resections., Methods: From January 1998 to December 2006, 300 patients underwent laparoscopic resection due to rectal carcinoma at our institution. We compared the converted patient group with the non-converted patient group regarding demographical, clinical, surgical, and histological data, compounded with the early and late postoperative results., Results: Two hundred seventy-four (91.3%) patients underwent laparoscopic rectal resection (LR), while conversion resection (CR) was necessary in 26 cases (8.6%). Conversion rate was 13% during the first 100 resections and decreased to 3% during the last 100 procedures (p = 0.035). Male gender, higher body mass index, and presence of T4-tumor were risk factors for conversion. Early postoperative complications were more frequent in the CR group than in the LR group. Concerning local tumor recurrence and overall survival, there was no significant difference between both groups (local recurrence, CR at 3.8% vs. LR at 4.5% and overall survival rate, CR at 76.9% vs. LR at 89.1%) after a median follow-up period of 22.5 months., Conclusion: Conversion to an open procedure during laparoscopic rectal resection correlates with an increased postoperative morbidity, however, without impairment of the short-term oncological outcome. The conversion rate is minimized by the growing experience of the operating surgeon and, therefore, is a marker of the learning curve.

Background: The video-assisted thoracoscopic approach has become the preferred method for many procedures due to the reduced trauma, complication rate and morbidity. The aim of this study was a risk evaluation of patients undergoing video-assisted thoracoscopic surgery (VATS) procedures., Methods: Between 1991 and 2004, 1,008 patients were included in this single-center retrospective analysis. Risk assessment was performed using univariate and multivariate analysis., Results: Multivariate analysis revealed that patient age (p = 0.003), the duration of the VATS procedure (p = 0.008), redo-VATS (p < 0.001) and conversion to open thoracotomy (p < 0.001) correlated significantly with the incidence of complications. Patients with immune deficiency following organ transplantation had the highest complication rate at 31.7%, which was significantly higher than for patients with either benign disease (p = 0.010) or malignant disease (p = 0.019)., Conclusions: VATS is a safe procedure, but extra caution is recommended for patients with a higher risk profile (age, redo-VATS, immune deficiency).

Background: Major abdominal operations were found to be associated with long-lasting metabolic changes, such as accelerated release of stress hormones and carbohydrate turnover. It is unknown currently whether acute changes of hepatic protein metabolism persist in a similar way. We wanted to determine the long-term dynamics of albumin synthesis and its relationship to whole body protein breakdown and albumin concentration after major rectal operations., Methods: We used stable isotope tracer techniques to determine albumin synthesis and whole body protein breakdown (rate of appearance of leucine, Ra) in postoperative patients about 1 week after low anterior rectal resection and also during convalescence (about 4 months after operation), and in healthy controls. Consecutive blood sampling was carried out during continuous isotope infusion (1-[(13)C]-leucine, 0.16 micromol/kg min)., Results: Serum albumin concentrations were close to the lower normal limit in patients early after operation but were comparable to controls in convalescent patients. Simultaneously, albumin synthesis was increased in the early postoperative phase (0.53 +/- 0.0.5%/h) compared with convalescent patients (0.32 +/- 0.04) and controls (0.28 +/- 0.04) (P < .01 each). A significant inverse correlation could be found between plasma albumin concentration and corresponding rates of albumin synthesis. Early after operation patients showed an increased leucine Ra (3.25 +/- 0.23 micromol/kg min) that was greater than that of convalescent patients (2.37 +/- 0.06 micromol/kg min, P < .05). Leucine Ra in both patient groups were greater than the rates in controls (2.01 +/- 0.07 micromol/kg min, P < .01) Albumin synthesis correlated weakly with whole body protein breakdown rate., Conclusions: Albumin synthesis and total body protein breakdown are increased after major abdominal operation, but albumin synthesis returns to control values only during convalescence. Hypoalbuminemia after rectal operations may be associated with high rates of albumin synthesis and is, therefore, not necessarily an indicator of insufficient hepatic function or poor nutritional status in that particular situation.

Background: Chronic allograft nephropathy (CAN) is, among others, caused by nephrotoxic side effects of calcineurin inhibitors (CNI), which are to date still the mainstay of immunosuppressive therapy. Sirolimus (SIR), an immunosuppressive compound without effects on glomerular perfusion, has been used in CNI-sparing immunosuppressive protocols. We report the 5-year follow-up of a prospective, controlled conversion study from CNI to SIR in patients with moderately to severely impaired renal function., Methods: Twelve renal transplant recipients with moderately to severely impaired renal function (estimated glomerular filtration rate of 17 to 35 mL/min according to the MDRD formula), enrolled in a prospective, controlled 1-year pilot study were followed for 5 years., Results: Three renal grafts (25%) were lost during the 5-year follow-up. Graft loss was due to noncompliance in one patient and to CAN in the other two patients. These two patients returned to dialysis 43 and 59 months after conversion, corresponding to 86 and 75 months after transplantation, respectively. Six of nine patients had a stable or even better renal function compared to the baseline. The lipid profile increased initially, but then remained stable over time., Conclusion: Conversion of immunosuppressive therapy from CNI to SIR in patients with impaired renal function more than 1 year after transplantation is feasible and safe yielding improved renal function in the majority of patients, which was sustained at 5 years follow-up.

Aim: Titanium-coated grafts for breast augmentation are available since 2001 and are used clinically. The titan surface is supposed to improve the tissue compatibility and to lower the infection rate. It was the aim of the present study to validate the antibacterial efficiency of titanium-coated silicone., Material and Methods: C3H/HcN mice were assigned to four different groups (n=6/group). Silicone without (group I and III) or with (group II and IV) titanium were implanted subcutaneously. Following this in groups III and IV a local contamination was induced with 2 x 10 (7) CFU/0.1 ml Staphylococcus aureus ATCC 25923. Groups I and II were not infected. 14 days after primary operation all animals were euthanized and the grafts harvested. Specimens were examined for signs of infections by macroscopy, histology and microbiology., Results: In group I none of the grafts were infected (0/5). In group II (silicone, + titanium, no contamination) one infection was evident due to biting of the animal (1/6). In group III (silicone, no titanium, contamination) an infection was detected in all mice (6/6). The use of titanium, however, did not significantly reduce the infection rate in contaminated animals (group IV, 5/6). Interestingly, tissue integration of titanium-coated grafts was macroscopically reduced compared to non- titanium-coated grafts (group II vs. I)., Conclusion: The titanium-coated silicone grafts were not effective in protecting infection in vivo. The decreased tissue integration of titanium-coated grafts, however, might reduce the rate of capsular contracture. This potential advantage of titanium needs to be validated in controlled clinical trials.

We report the case of a patient who developed an asymptomatic pseudoaneurysm in the left external iliac artery after transplant nephrectomy. The pseudoaneurysm most probably arose as a suture aneurysm from the external iliac artery after removal of the graft renal artery. Obviously we can not exclude the possibility it was a true aneurysm, although this seems much less likely. The pseudoaneurysm was detected during a routine CT scan and was treated interventionally with a stent-graft. One month later the asymptomatic patient underwent a vascular ultrasound examination including color Doppler, power Doppler, and B-flow as a routine control. An endoleak with collapse of the stent-graft was diagnosed. There was no evidence of stent infection. At a reintervention, the pseudoaneurysm was successfully treated using two uncovered Palmaz stents at the proximal and distal edge of the stent graft. Peri- and post-interventional ultrasound and CT angiography confirmed the exclusion of the aneurysm without an endoleak.

Appropriate nutritional therapy of surgical patients intends to supply calories for the maintenance of essential body functions. Beyond this goal, nutritional support may also significantly reduce nosocomial morbidity if applied properly and to the right patients. In surgical patients, nutritional therapy should start preoperatively by identifying and treating malnutrition and be continued postoperatively as a patient-tailored supportive measure. Oral/enteral nutrition is feasible in the majority of patients. Rare exceptions are patients with intestinal leakage, overt ileus, and circulatory shock. If the upper gastrointestinal tract is not functioning (as in swallowing disorders or after construction of surgical anastomoses), tube systems may be used. They can be placed endoscopically or at the time of surgery (needle catheter jejunostomy) to allow continuous enteral nutrition. If oral/enteral nutrition cannot completely meet caloric requirements of the patient, additional parenteral supply is indispensable to reach the intended caloric goal.

Hyperglycemic hyperinsulinemia cannot stimulate intestinal protein synthesis in healthy individuals but does so in conditions characterized by an altered somatotropic axis such as diabetes. Only in a state of growth hormone resistance (high growth hormone but low insulin like growth factor [IGF-1] concentrations), extra insulin may acutely reverse the impaired, growth-hormone-induced IGF-1 release, thereby exerting anabolic actions at the intestinal tract. Growth hormone resistance can be also found in patients after surgical stress. Therefore, we wanted to test the hypothesis whether hyperglycemic hyperinsulinemia would stimulate ileal protein synthesis in the latter condition. Mass spectrometry techniques (capillary gas chromatography/combustion isotope ratio mass spectrometry) were used to directly determine the incorporation rate of 1-[(13)C]-leucine into ileal mucosal protein. All subjects had an ileostomy, which allowed easy access to the ileal mucosa, and consecutive sampling from the same tissue was performed during continuous isotope infusion (0.16 mumol/kg min). Isotopic enrichments and fractional protein synthesis were determined at baseline (period I) and after a 4-hour glucose infusion (170 mg/kg/h) or after infusion of saline (control group) (period II). In controls, ileal protein synthesis declined significantly during prolonged isotope infusion (period I: 1.11 +/- 0.14%/h, period II: 0.39 +/- 0.13%/h, p < .01). In contrast, ileal protein synthesis remained constant during glucose infusion (period I: 1.32 +/- 0.35%/h, period II: 1.33 +/- 0.21%/h, n.s. vs period I, but p < .005 vs the corresponding value at the end of period II in the control group). Using the continuous tracer infusion technique, ileal protein synthesis seemingly declines over a short time in control subjects. We found evidence that this artificial decline was due to mass effects of a rapidly turning over mucosa protein pool in which an isotopic plateau was reached during the experiment and of which the size amounted to approximately 4% of the total mixed protein pool. Maintenance of ileal protein synthesis during glucose infusion therefore indicates a rise of ileal protein synthesis in a slowly turning over protein pool. This effect in postsurgical patients would be compatible with the concept of intestinal insulin action to depend on the specific clinical state (eg, growth hormone resistance).

Objective: Peripheral blood mononuclear cell (PBMC) dysfunction occurs following major abdominal surgery and correlates with an increased rate of septic complications. Studies have shown that dehydroepiandrosterone (DHEA) restores cell-mediated immune responses after trauma-hemorrhage in mice. Nonetheless, it remains unknown whether DHEA has any salutary effects on depressed PBMC function in surgical patients., Design: Laboratory experiment., Setting: University laboratory., Patients: Fifteen patients undergoing major abdominal surgery., Interventions: Blood samples were obtained preoperatively and 2 hrs postoperatively., Measurements and Main Results: PBMCs were cultured with 33% plasma in the presence or absence of DHEA (10(-10) M, 10(-8) M physiologic concentration, 10(-6) M, 10(-5) M). In an additional set of samples, the estrogen receptor antagonist tamoxifen (10(-6) M) was added. The release of proinflammatory cytokines (interleukin-1beta, interleukin-6, and tumor necrosis factor-alpha) was measured in the supernatants by enzyme-linked immunosorbent assay. Abdominal surgery resulted in depressed interleukin-1beta and tumor necrosis factor-alpha release by PBMC. Addition of DHEA to the culture medium, however, significantly improved the release of interleukin-1beta and tumor necrosis factor-alpha and stimulated the interleukin-6 release capacity of PBMC. This effect was most pronounced for a concentration of 10(-5)M DHEA. The immunomodulatory effect of DHEA on PBMC cytokine release was completely blocked by tamoxifen. In contrast, the modulatory effect of DHEA was enhanced by the addition of postoperative plasma., Conclusions: DHEA stimulates proinflammatory cytokine release capacities of human PBMCs following major abdominal surgery. The estrogen receptor appears to be involved in mediating the immunomodulatory effect of DHEA. Thus, DHEA might be a useful adjunct for preventing immunosuppression in surgical patients.

Hypothesis: Laparoscopy has become a popular approach for the surgical treatment of benign and even malignant colorectal diseases. Several authors have reported better preserved immunity in patients undergoing laparoscopic compared with conventional colorectal surgery. The present study addresses the hypothesis that specific and nonspecific immunity are differently affected by laparoscopic and conventional colorectal surgery., Design: Nonrandomized control trial., Setting: University hospital., Patients: Seventy prospectively enrolled patients with colorectal diseases undergoing laparoscopic (n = 35) or open (n = 35) surgery., Main Outcome Measures: Blood samples were taken prior to surgery as well as on days 1, 3, and 5 after surgery. Cell numbers of lymphocyte subpopulations as well as natural killer cells were determined by flow cytometry, and interleukin 6 and C-reactive protein serum levels were measured., Results: Significant differences between study and control patients (P<.05) were detected regarding circulating interleukin 6 and C-reactive protein levels with a reduced proinflammatory response to surgery in patients after laparoscopic surgery. Furthermore, postoperative natural killer cell counts were significantly higher in patients after laparoscopic surgery. The levels of B lymphocytes and T lymphocytes and helper T-cell counts and cytotoxic (suppressor) T-cell counts did not show significant differences after open or laparoscopic surgery., Conclusions: Our findings indicate a less pronounced proinflammatory response to surgical trauma in patients after minimally invasive surgery. The nonspecific immune response appears to be less affected by laparoscopic surgery when compared with open surgery while the specific cell-mediated immunity is equally affected. These findings are important because a divergent effect on specific and nonspecific immunity of laparoscopic surgery for colorectal disease has not been reported before.

Introduction: Tension-free incisional hernia repair using alloplastic material increasingly replaces conventional repair techniques. The aim of the present study was to evaluate the early and long-term complications as well as patients' satisfaction., Methods: Laparoscopic hernia repair with intraperitoneal mesh implantation (PTFE) was performed on 28 patients at the Klinikum Grosshadern between 2000 and 2003 (16 males, 12 females, average age 61.2). Intra- and postoperative complications were registered retrospectively. In addition, 25 patients were evaluated for recurrence, postoperative pain and patient contentment (median follow-up 383 days)., Results: A low complication rate was observed in our patient collective. One trocar bleeding occurred. 2 patients presented with wound hematoma. The recurrence rate was 8 % (2/25). 60 % of the patients were free of pain postoperatively. 88 % would once again choose the laparoscopic approach for incisional hernia repair., Discussion: The laparoscopic technique was associated with a low recurrence rate, a small rate of wound infections and high patient comfort. Thus, the laparoscopic approach for mesh implantation appears to be a safe and effective method for the treatment of incisional hernias. The efficiency of laparoscopic intraperitoneal mesh implantation, however, should be evaluated within a prospectively randomized multicenter trial.

Background: There is a high incidence of postoperative complications and late recurrences after operative therapy of a pilonidal sinus. The optimal treatment strategy is still matter of discussion. We studied the long-term results after excision of a pilonidal sinus and primary midline closure compared with the open surgical procedure., Materials and Methods: A total of 73 patients (62 male and 11 female, mean age 26.6 years) underwent a total of 79 operations between 1992 and 2001. Thirty patients (38%) were previously operated on because of a pilonidal sinus.Twenty-four were treated in our institution by an open procedure (five after simple abscess incision, 19 after sinus excision) and 52 by primary midline closure. Another three patients received skin flap procedures., Results: Follow-up was possible for 65 patients (82%) for a median of 50 months.Recurrent pilonidal sinus occurred in 22 cases: 18 after primary midline closure (42%) and four after open procedure (21%, P=0.4). We found no relapse following the three skin flap procedures. There was a significantly higher relapse rate in patients operated with recurrent disease (12/25 vs 10/40; P<0.05)., Conclusions: Despite of numerous previously operated patients (38%), there was a high recurrence rate (42%) after excision of a pilonidal sinus and primary midline closure. Alternative operative techniques creating a lateral wound or the various skin flap procedures may be promising alternatives. We are in the process of changing our treatment strategy for patients suffering from a pilonidal sinus.

Background: Biliary tract lesions pose a dreaded complication of laparoscopic cholecystectomy. In a retrospective study we analyzed the clinical presentation, diagnostic and therapeutic management and outcome of 28 patients presenting with iatrogenic bile duct injuries., Patients and Methods: Between 1994 and 2001 we treated 28 patients with bile duct lesions following laparoscopic cholecystectomy at our center. Operation notes and charts of all patients were reviewed systematically. A follow-up examination of each patient was performed after a median of 12 months (range 1-90)., Results: Twenty-two patients presented with major circumferential bile duct defect lesions. Less severe injuries (n=6) were two minor bile leaks, one bile duct stricture and three tangential lesions. Twenty-six patients were referred to our institution within 16 days (range 0-226 days). Six patients were treated by nonsurgical procedures: endoscopic stenting in four and percutaneous intervention in two. In one of the remaining patients a cystic duct leak was closed via laparotomy, and in 21 a hepaticojejunostomy was performed. Reconstruction of a hepaticojenunostomy was performed in two of these patients. Patients were dismissed from the hospital after a median of 13 days (range 4-156). Four patients presenting with generalized biliary peritonitis required prolonged intensive care. One or more episodes of cholangitis were seen in five patients during follow-up examinations., Conclusions: Major iatrogenic bile duct injuries are associated with high morbidity and prolonged hospitalization. Interdisciplinary cooperation and early referral to an experienced center is crucial in the management of patients suffering from this affliction. Cholangitis is a marked problem in the follow-up.

Background: Major liver resection is a routine surgical treatment, but hemodynamic and pulmonary complications are common. We investigated the effects of hepatic resection on hemodynamics and pulmonary and liver function., Patients and Measurements: Twelve patients who underwent major liver resection due to primary liver tumor, liver metastasis, or hemangioma were investigated prospectively and consecutively. Six patients who underwent gastrectomy due to gastric cancer served as a control group. Hemodynamic parameters (cardiac index, intrathoracic blood volume, mean arterial blood pressure), extravascular lung water, and indocyanine green clearance were measured after the induction of general anesthesia, during the preparation period, during total hilar clamping ("Pringle's maneuver"), at the end of surgery, and 24 and 72 h after surgery., Results: In contrast to gastrectomy, patients who underwent liver resection developed a hyperdynamic circulatory state (cardiac index 72 h postoperatively: 4.6+/-1.2 l/m(2) vs. 3.6+/-0.6 l/m(2)). Simultaneously we observed a significant increase in extravascular lung water in the liver resection group, indicating a moderate pulmonary edema. Indocyanine green clearance did not deteriorate following liver resection., Conclusions: After liver resection the physiological response resulted in elevated cardiac output and moderate pulmonary edema. Dynamic liver function was elevated within 24 h due to an increase in perfusion and regeneration activity of the remaining parenchyma.

Introduction: In 1993, Le Gall proposed a new Simplified Acute Physiology Score (SAPS II) to assess the severity of illness in intensive care patients and to predict the risk of hospital mortality using a large data base of more than 13,000 patients of different intensive care units. Up to the present time, no satisfactory form of validation for surgical intensive care patients has been available. We investigated the prognostic quality of this score system for the assessment of the severity of the illness in surgical intensive care patients., Methods: Between October 1997 and December 1998, 310 consecutive patients were included in our study. We calculated the SAPS II score according to the published guidelines. Additionally, we calculated the risk of mortality by referring to the score. We used a receiver operating characteristics (ROC) analysis for statistical evaluation., Results: The median duration of stay on our intensive care unit was 8.4 days. In all patients, the mean SAPS II was 29.9 +/- 12.7. In the survivors, calculation of mean SAPS II was 27.7 +/- 11.4. In the non-survivors, mean SAPS II was 45.7 +/- 11.2. The predicted risk of mortality was 15.7%, the actual risk of mortality was 15.8%., Conclusion: The SAPS II score has proved to be a good prognostic instrument in surgical intensive care patients. SAPS II is particularly effective in providing an exact estimation of the risk of death, classifying patient groups in clinical studies, epidemiological investigations, and quality assessment.

Malnutrition is correlated with an increased incidence of postoperative complications. As a consequence, preoperative hyperalimentation results in a decrease of postoperative complications in patients with malnutrition. In contrast, the beneficial effects of postoperative nutritional support are still to be proven. The majority of prospective trials have shown limited advantageous effects for patients with delayed nutritional intake after the surgical procedure. In these patients enteral nutrition has been shown to reduce the rate of infectious complications. If enteral nutrition cannot be applied, the addition of glutamine as a dipeptide is beneficial with regard to intestinal barrier function.

Functional variables after coloanal anastomosis or anastomosis with J pouch were investigated in 40 patients in a prospective randomized study. Continence for liquids and gas control were superior after J pouch anastomosis compared with coloanal reconstruction. The neorectal capacity was higher after J pouch anastomosis. The perception threshold for stool filling was higher in patients with J pouch resembling those values observed preoperatively. Quality of life was improved in subjects with with J pouch, although differences did not reach a significant level.

Trauma, operative interventions, infection and other disturbances of homeostasis lead to a uniform reaction of the body, namely release and activation of hormones and cytokines. Profound alterations of substrate flow result, with mobilization of energy stores and degradation of structural and functional proteins of vital organs like the gut mucosa. Due to these reactions the energy demands of the organs are met and energy-consuming synthesis of substrates is indicated. Clinically, hypermetabolism, hyperglycemia, lipolysis and increased urea production with negative nitrogen balance can be observed. The metabolic reactivity is reached by an increased substrate cycling. To avoid negative consequences such as organ dysfunction, a rational situation-adapted substrate supply is warranted as well as reduction of catabolic stimuli and stimulation of anabolic factors. The metabolic care of the surgical patient is still a basic and important task.

Disturbances of the arterial perfusion of the graft following liver transplantation (LTx) are mainly of technical origin and contribute considerably to the postoperative morbidity and lethality. Aim of this retrospective survey was to determine the incidence and the consequences of hepatic artery thrombosis (HAT) and stenosis (HAS) in 203 patients (pts.) who underwent 246 liver transplantations. HAT was identified in 22 pts. by clinical, laboratory and sonographic and/or angiographic means. In 4 pts. HAT was asymptomatic and was detected during routine examination. Two pts. underwent immediate retransplantation, of which only one survived. Primary revascularization was performed in 16 pts., but was successfully only in 5 pts. Six pts. eventually underwent retransplantation with only one survivor. Biliary complications after HAT were observed in 10 pts. (45%), presenting in 8 pts. as biliary leak, in 2 pts. as bile duct stenosis. In the 12 patients with HAT who died the leading causes of death were sepsis (n = 3) and multiple organ failure (n = 3). HAS was observed in 11 pts., of which one was asymptomatic. Ten pts. underwent surgical revision. Redo of the arterial anastomosis was the most common procedure. Four pts. survived long term. Biliary tract complications were seen in 3 pts. (leak 1, stenosis 2). Three pts. died secondary to HAS from sepsis. These results confirm the life threatening character of any arterial complication after LTx. Because other reasons are rarely detected, the majority is attributable to technical faults. In order to avoid arterial complications extraordinary care has to be taken in the surgical handling of the arterial supply of the graft during harvesting, back table work and transplantation.

Objective: Selection of important results in pathophysiology and pathobiochemistry of the surgical patient., Sources: Anglo-American and German literature since 1988., Selection Criteria: Importance of reported results for the understanding of superior interactions., Results: The term 'postaggression' syndrome was introduced on the basis of earlier observations regarding the pathophysiology of the surgical patient. More recent results which addressed pathobiochemistry, and which had been obtained by methods of molecular biology, led to the new definition as systemic inflammatory response syndrome (SIRS). Knowledge on metabolic changes which are associated with SIRS increased significantly in the field of protein and carbohydrate metabolism. Concerning the clinically important phenomenon of protein catabolism, it is possible today to differentiate between changes in protein synthesis and protein degradation. The time period which had passed since the underlying trauma and the variable responses of different organ systems were recognized as important variables of catabolism. The ongoing question regarding possible mediators of protein catabolism still remains unanswered. Release of cytokines which can be observed in SIRS appears to play only an indirect role. The small bowel, which is important for the general pathophysiology of SIRS, gained a central position in amino acid metabolism (especially glutamine metabolism). A new aspect of carbohydrate metabolism was found in the liver. The accelerated cycling of glucose molecules between glucose, glucose-6-phosphate, and glucose (glucose cycling) seems to contribute to the increased energy expenditure found in SIRS., Conclusions: Use of new methods in in vivo and in vitro research significantly expanded the knowledge on pathophysiological and pathobiochemical mechanisms in SIRS. Future research activities should tie together these individual mechanisms into the complex network as it presents to the physician during clinical routine.

Homologous blood transfusion has been associated with an increased risk of postoperative infectious complications. To test the clinical consequences of this apparently immunosuppressive effect of homologous blood in a controlled trial, we designed a study in which the control group deposited autologous blood before their operations for use should transfusion be needed. We enrolled 120 patients with apparently curable colorectal cancer who were able to predeposit autologous blood (haemoglobin > 12.5 g/dL). 62 patients were assigned to receive homologous blood if blood transfusions were needed during operation, and the other 58 to receive their own predeposited blood followed, if necessary, by homologous blood [corrected]. Despite the similarity between the groups in factors known to affect the risk of postoperative infections, there was a significant difference in postoperative infection rate between the homologous and autologous blood groups (17 [27%] vs 7 [12%], p < 0.05; unadjusted odds ratio 2.75 [95% CI 1.07-7.11). The rates of non-infectious complications were similar Probably because their preoperative blood depositing caused the autologous blood patients to have lower haemoglobin concentrations, they were more likely to require transfusion than were the homologous blood group (53 [91%] vs 37 [60%], p < 0.001; relative risk 1.53 [1.24-1.89]). 20 (35%) required homologous as well as autologous blood. To adjust for the many infection-related factors, we did multivariate regression analysis; tumour location, preoperative ASA index, and study group assignment were the only significant risk factors. The odds ratio for postoperative infections adjusted for these factors was 2.84 (1.02-7.98, homologous vs autologous). Testing of delayed-type hypersensitivity responses before and after surgery showed decreases in both mean diameter and number of positive reactions in recipients of homologous blood and slight increases in those who received autologous blood. This study shows the clinical potential of blood-transfusion-mediated immunomodulation, which may be important also in tumour immunology.

Extended experimental studies revealed different immunological mechanisms which are possibly responsible for blood transfusion-associated immunosuppressive conditions. To expect a clinical impact of these mechanisms on the course of tumor disease, it is necessary to postulate (1) that immunological mechanisms have a significant role in controlling tumor growth and (2) that blood transfusion-induced immunmodulation is long lasting. Both postulates are supported by recent reports and are the rationales of clinical studies indicating that blood transfusion is a risk factor for postoperative infections and tumor recurrence. Since all studies have been retrospective or uncontrolled, we performed a prospective controlled study in randomized groups of patients suffering from colorectal cancer and compared the effects of allogeneic and autologous blood transfusions. The results indicate that patients treated with allogeneic blood transfusion had significantly higher rates of postoperative infectious complications than patients who received autologous blood. Our preliminary follow-up observations found a trend towards higher tumor-free survival in patients treated with autologous blood which is statistically significant in subgroup analysis.

In seven healthy patients we studied the influence of various doses of insulin on the substrate concentration of free fatty acids and ketone bodies on the first day after major abdominal surgery by the glucose clamp technique. Insulin was infused at a rate of 0.2 and 1.0 mU/kg bw.min for ninety minutes and thus insulin concentrations were raised twice and seven times above the basal value. Already the minimal infusion of 0.2 mU insulin/kg.bw.min lead to a significant decrease of free fatty acids and ketone bodies. Our results show that even in postoperative stress the lipolysis of adipose tissue can be reduced by minimal insulin doses.