Preoperative Glucocorticoid Use in Major Hepatectomy

Overview

Background: Reducing postoperative complications remains a dominant challenge for all clinicians. By minimizing the incidence of adverse outcomes, health care costs and patient recovery can be improved. A number of studies have documented that fewer postoperative adverse events occur with the preoperative use of glucocorticoids (GC). Two small manuscripts support the use of GC with particular reference to liver resection. Major hepatectomy can be associated with numerous adverse outcomes (hemorrhage, bile leak, liver failure, wound infection other infectious complications). This rate approaches 54% in some trials. Additionally, the liver is responsible for numerous metabolic functions and actively participates in the acute phase response via the generation of inflammatory mediators and cytokines. Glucocorticoids play an important role in suppressing the over-synthesis of pro-inflammatory cytokines and therefore may assist in reducing postoperative complications. These pro-inflammatory cytokines recruit leukocytes at the site of injury, leading to hepatic parenchymal cell damage. Decreasing pro-inflammatory cytokine ratio may therefore reduce hepatic damage. Methods: The investigators will complete a randomized controlled trial evaluating the incidence of postoperative complications and overall hospital length of stay in adult patients following preoperative vs. no administration of 500 mg of methylprednisolone (MP) within 30 minutes of starting a hepatectomy. This trial will employ block randomization and intention to treat protocols. Four HPB surgeons at University of Calgary will participate in the trial with a planned inclusion of all patients scheduled to undergo hepatectomy in a period of 2 years. Analysis: Standard statistical analysis will include normally or near-normally distributed variables reported as means and non-normally distributed variables as medians. Means will be compared using the student's t test and medians using the Mann-Whitney U test. Differences in proportions among categorical data will be assessed using Fischer's exact test. A p value \< 0.05 will represent statistical significance for all comparisons. Hypothesis: Preoperative GC administration will decrease the incidence of postoperative complications and overall hospital length of stay following hepatectomy. Potential Impact: Health care utilization and economics are an expanding area of importance. Reducing post-hepatectomy complications are crucial to this end.

Introduction: Major hepatectomy lead to significant postoperative anatomical and physiological changes. The liver dictates the systemic metabolic environment and helps maintain homeostasis. When segments of liver are removed, it is challenged to regenerate and restore cellular volume, as well as continuing to perform mandated metabolic functions. Major hepatectomy can result in multiple complications (hemorrhage, biliary fistulae, wound infection, liver failure) with an overall complication rate up to 54% in some studies. Recent studies purport that preoperative administration of GC may suppress the acute phase response during the early postoperative period. GCs are steroid hormones with a variety of immunological and biochemical functions (anti-inflammatory attenuation of the acute phase response by binding to intra-cytoplasmic GC receptors and therefore increased transcription of anti-inflammatory proteins interleukin-10 (IL-10), IL-1-receptor antagonist (IL-Ra), lipocortin-1 and neutral endopeptidase, or direct inhibition of activated transcription factors (nuclear factor kB, activator protein-1)). GC can also suppress gene expression by altering the chromatin structure and tightening DNA coils. The acute phase response is a neuro-humoral and time-limited defensive mechanism, which aims to restore homeostasis. It is characterized by fever, leukocytosis, gluconeogenesis, as well as changes in lipid, carbohydrate and protein metabolism. Activation of the hypothalamus-pituitary-adrenal axis, coagulation and compliment pathways is also one of the salient features of the acute phase response. It is regulated by cytokine receptor antagonists (IL-1Ra) and soluble cytokine receptors. Cytokines are soluble, low molecular weight glycoproteins that act as inflammatory mediators and can be broadly divided into pro and anti-inflammatory cytokines. At low concentrations, they have a paracrine effect, while in high concentrations they act systemically. These pro-inflammatory cytokines (IL-1, IL-2, IL-6 and tumor necrosis factor (TNF) alpha) remain in the plasma for at least 24 hours after generation. They also attract leukocytes (neutrophils and macrophages) to the site of injury and are detrimental if produced in large amounts. Oka et al have demonstrated that increasing levels of IL-6 are directly related to an increased rate of postoperative complications. In low doses, GC reduce short term mortality, reverse the shock process and improve hemodynamics in septic shock patients. In a meta-analysis of 17 RCTs, Annane et al found that 28-day mortality for a low-dose dexamethasone (DXA) group, versus control group was 35.3% vs. 38.5%. Shock reversal was also increased in the DXA group (66.9% vs. 58.6%) and the ICU stay was decreased by 4.49 days. Preoperative use of GC improves cardiac functions after coronary artery bypass grafting. The likely underlying mechanism for this cardio-protective effect is reduced release of pro-inflammatory cytokines IL-6, IL-8, and TNF-alpha with a concurrent increased release of anti-inflammatory cytokine IL-10. In cases of laparoscopic cholecystectomy, 50% to 75% of patients express nausea and vomiting in the early postoperative period. In a meta-analysis, Karanicolas et al found that the absolute risk of nausea in low risk patients. Similarly DXA decreases the absolute risk of vomiting from 12% to 5% in low risk and 30% to 12% in high risk groups as compared to placebo. Although the quality of evidence is low regarding postoperative pain in post-cholecystectomy patients, researchers have found less postoperative pain in the DXA group. Gomez-Hernandez et al also suggested that a preoperative single dose of DXA decreased postoperative nausea, vomiting and pain in patients undergoing mastectomy (28.6% vs. 60% in the early postoperative period). The patients in placebo group required more anti-emetic medications. DXA reduced postoperative pain significantly and fewer patients requested analgesia (10 vs. 21). Additionally there was no increase in adverse events, morbidity and mortality in the steroid group. Preoperative use of DXA significantly improves postoperative pain, nausea, vomiting and vocal function within the first 48 hours after thyroid surgery. Subclinical renal dysfunction is frequently noted following liver transplantation. This occurs mainly due to hepatic ischemia/reperfusion injury. Turner et al suggested that GC decrease the incidence of renal tubular dysfunction when given at the time of induction of anesthesia. The main difference in serum creatinine level was noted on postoperative day 3 (164.8 mumol/L in the saline group vs. 88.5 mumol/L in the MP group). Additionally, patients in the saline group suffered from more complications (47% vs. 12%). Studies have suggested that GC limit the peritoneal inflammatory response and therefore help in recovery and fatigue in the early postoperative period. Low levels of IL-6 and IL-13 were found in the peritoneal fluid of patients receiving GC on postoperative day 1. Schulze et al observed improved pulmonary function and early mobilization in a MP group, as compared to placebo, in colon surgery patients. Furthermore, they noted a significant decrease in IL-6, C reactive protein and pain scores. The appetite was increased significantly in patients on GC, however there was no difference in wound healing and other infectious complications (including anastomotic leaks). In colorectal surgery, GC do not appear to increase postoperative complications. A double blind, randomized clinical trial consisting of 30 patients undergoing colonic surgery, showed that there was no significant difference in postoperative IL-6, C reactive protein, pain scores, bowel functions, mobilization, hospital length of stay, complications and readmission rates between the two groups when a single 8 mg preoperative dose of DXA was given to the interventional group. A recent systematic review and meta-analysis of 11 moderate quality RCTs suggested decreased postoperative complications and hospital length of stay of patients undergoing major abdominal surgery following preoperative use of GC. In the case of liver resection surgery, Aldrighetti et al found decreased serum ALT, AST, total bilirubin, IL-6, TNF-alpha, hospital length of stay and postoperative complications following a single preoperative GC dose. The dominant clinical rationale for not using preoperative steroids is to avoid related side effects. A substantial number of studies have documented fewer adverse events in MP groups as compared to placebo, suggesting there is a need to reconsider the steroid-free protocol in the preoperative period. Further appropriately designed and well-powered studies are urgently needed to evaluate the effects of preoperative GC use. As a result, we have planned a prospective RCT at the University of Calgary to determine the role of GC in reducing postoperative complications and hospital length of stay for patients undergoing partial hepatectomy. Recruitment: Contact information for study candidates will be obtained by the project's research staff from the offices of the four participating surgeons. Patients will be contacted prior to surgery by the research staff and will be provided information regarding the study. Patients who consent to participate will be randomized (the clinic at the time of consent completion and operative scheduling). Data collection: Patients will be tracked postoperatively for the development of complications and overall length of hospital stay. Information will be collected by chart review from the operative report and patient record by research staff. Data elements to be collected will include, but not be limited to: * Date of birth, Gender * Surgeon, procedure, surgery duration * Antibiotic prophylaxis * Estimated blood loss, Transfusion * Wound closure technique, peroperative drain, Wound infection or disruption * Hepatic failure * Acute kidney injury * Intraperitoneal abscess, surgical site infection * Pneumonia, Blood stream infection, Pulmonary embolism, Deep venous thrombosis * Hypoglycemic incidents * Bile leak (requiring percutaneous drainage or via an operatively placed drain) * Anastomotic gastrointestinal leaks (combined colorectal / hepatic resection) * Preparation type (ETOH based, chlorhexidine or iodine based) * Time to oral intake Patient records: * Preoperative chemotherapy * Preoperative albumin level * Preoperative PVE and /or TACE * Underlying liver disease (cirrhosis) * ASA score * Length (if any) of hepatic inflow occlusion, * Length of hospital stay \& mortality * Co-morbidities (smoking, diabetes, hypertension, COPD, Ischemic heart disease, arrhythmias, metastatic cancer, AIDS, obesity, alcohol abuse,psychosis, anemias etc.) * Glucocorticoids - methylprednisolone given, dose, route of administration and time before starting surgery. The investigators will also track IL-6 and C-Reactive protein (CRP) levels on postoperative day 1 (with daily blood work) as a marker of global immune function. Definition of complications can be found in appendix 2. Power calculation to determine sample size: The largest surgical trial to date recruited a total of 73 patients. There are 5 small hepatic resection RCTs available for analysis (all trials utilized MP). The largest of these studies noted a decrease in overall complications from 54% to 14% with GC use. With a consistent volume of hepatectomies (100+ per year at FMC), and a planned inclusion/capture for our group's resections of nearly 100%, we expect to randomize over 100 patients in each arm in 2 full years. With an alpha = 0.5, probability = 80%, predicted effect size of 0.30 (2-tailed test), the sample size required would be 84 patients. Additional comments: Methylprednisolone (MP) is the best choice of GC because it: (1) has a 5-fold higher anti-inflammatory action compared to cortisol, (2) has reduced effects on electrolytes, (3) duration of MP biologic action is 36 hrs (half life = 2.8 hrs) so both phases of ischemia-reperfusion injury will be covered, and (4) 500 mg of MP results in a defined and predictable response of \>1 ug/ml in blood and \>10 ug/ml in liver tissue. Resources: This project is built on an existing research team who work collaboratively with the Principal Investigator to carry out the project effectively and efficiently. An experienced research assistant will assist in data collection, analysis, preparing reports for publications and organizing research meetings. Appendix 2: Hepatic failure; Hyperacute and/or acute onset of coagulopathy (INR\>1.5), reduction in synthetic function, and encephalopathy requiring therapy. Acute kidney injury; An abrupt (\<48 hrs) reduction in kidney function (absolute increase in serum creatinine \> or equal to 26.4 umol/L, a \> 50% increase in serum creatinine or a reduction in urine output of less than 0.5 ml/kg per hour for more than six hours. Intraperitoneal abscess; Identified on cross-sectional imaging. Bile leak; Either requiring percutaneous drainage or evidence of bile in an operatively placed closed suction drain. Wound disruption, dehiscence or hernia Pneumonia; Blood stream infection; Urinary tract infection; Surgical site infection. Anastomotic gastrointestinal leaks (combined colorectal/hepatic resections); GI leak confirmed by imaging (fistulogram, CT) and requiring intervention. Deep venous thrombosis; Identified on ultrasonography or cross sectional imaging prompted by clinician suspicion or incidental imaging. Pulmonary embolus; Identified on cross-sectional or V/Q scan imaging. Hypoglycemic incidents; Blood glucose \<3.0 mm/L with concurrent patient symptoms of hypoglycemia.