Topical Vancomycin Over Sternal Edge in Cardiac Surgery

Overview

This is a single-blind 1:1 randomized controlled trial based on the hypothesis that topical application of vancomycin paste over sternum edge is safe and can reduce sternal wound infection after elective cardiac surgery. Vancomycin paste will be prepared using 2.5 g of vancomycin mixed with 2 ml normal saline. Vancomycin paste as control or 2 ml normal saline as placebo will be spread on sternal edge immediately after sternotomy and before sternal closure. The safety of Vancomycin paste over sternum edge will be assessed according to postoperative serum Vancomycin exposure and potential side effects such as renal toxicity or bacterial resistance. Effect of topical Vancomycin on incidence of postoperative sternal wound infection will be assessed on postoperative 7, 30, and 90 days.

1\. Patient Recruitment and Group Assignment 1. Institution. All the surgical procedures will be undertaken in the National Cheng Kung University Hospital (NCKUH) under service of the Cardiovascular Surgery Department. The patient monitoring will be carried out perioperatively in the general ward and intensive surgical unit (ICU, surgical or cardiac) of National Cheng Kung University Hospital. 2. Eligibility Criteria. Described in other section. 3. Randomization. Eligible patients will be 1:1 randomized using a computer-generated list to receive either normal saline (Control group) or Vancomycin Paste (Vancomycin group). A randomization list with corresponding patient recruitment number will be archived in NCKUH pharmacy. On the operating day, our research coordinator will order a study prescription, which looks identical to either control or Vancomycin group. The pharmacy will deliver the prescription content according to the patient recruitment number on the randomization list. The surgeons will know the patient grouping only after they get the prescription content in the operating room. 4. Blinding. The intraoperative staff including the surgeon will know the application material for the patients. The patient and the nursing staff in charge of sternal wound evaluation will not know the allocation of patients. 5. Sample Size Determination. The average incidence of DSWI after isolated CABG in NCKUH is 3-4% in recent 5 years. An expected reduction of DSWI from 4% to 1% leads to an effect size of 0.15, which will require a total sample size of 330 to achieve a statistical power of 0.8 given an alpha error probability of 0.05 and 1 degree of freedom. (Estimation with G\*Power software). Considering the potential patient withdrawals, our planned sample collection will be 360 cases (180 in each group) in two years. \------------------------------------------------------------------------------------------------------- 2\. Intervention (1) Vancomycin group protocol. i. Timing of application: the Vancomycin® (China Chemical \& Pharmaceutical Co., Ltd. (CCPC / Taiwab) paste will be spread on sternal edge immediately after sternotomy and before sternal closure. ii. Regimen: The vancomycin paste will be prepared using 2.5g of vancomycin mixed with 2ml normal saline for each time. A total of 5gm of vancomycin® powder will be applied. (2) Control group protocol. 2ml normal saline will be spread on sternal edge immediately after sternotomy and before sternal closure. (3) Common perioperative protocols in both groups regarding sternal wound infection prevention: i. Intravenous antibiotics: Cefazolin 1g q8h for 24hrs. Patients with known cephalosporin or beta-lactam allergy will receive single dose Vancomycin plus Gentamicin. ii. Skin preparation: Skin shave will be performed on the operation day after injection of prophylactic antibiotics. Skin sterilization will be prepared with alcohol beta-iodine scrub followed by Hibitane (Chlorhexidine Gluconate). iii. Bone wax: bone wax usage for hemostasis will be minimized and be removed as much as possible before sternal closure. iv. Internal mammary artery harvest: Bilateral internal mammary arteries will be used at the discretion of the individual surgeon. v. Sternum closure: our standard is 8 single wires closure. Replacement of any 2 single wires with a figure of eight wire or addition of Robicsek technique will depend on the decision of the individual surgeon. vi. Subcutaneous and skin closure: interrupted Vircyl for subcutaneous closure and staples for skin. vii. Blood glucose control: intravenous insulin infusions to maintain serum blood glucose level between 120 and 180 mg/dL. viii. Chest hugger: postoperative chest hugger will be used according to the preference of the individual patient. ix. Delayed sternal closure: The decision is based on the discretion of the individual surgeon. If a delayed closure is attempted, the vancomycin paste will still be applied over the unclosed sternal edge. \------------------------------------------------------------------------------------------------------- 3\. Outcome Measurement. Details described in other section. (1) Sternal wound. Measurement on postoperative day (POD) 7, 30, and 90, based on the definition from Centers for Disease Control and Prevention. (2) Serum Vancomycin levels. Sampling time: postoperative day 0, 3 and 7. Measurement: High performance liquid chromatography (HPLC). \------------------------------------------------------------------------------------------------------- 4\. Data and Registry. 1. Data dictionary. Each variable in a row, presented in the following order: Variable, Coding Information, Range (numeric data) or Levels (categorial data), Information Source i. Sex, Sex, M or F, Electrical Medical Record ii. Age, Age, 20 - 100, Electrical Medical Record iii. Height, Body Height, 100 - 200, Electrical Medical Record iv. Weight, Body Weight, 30 - 150, Electrical Medical Record v. DM, Diabetes Mellitus, 1 or 0, Electrical Medical Record vi. DM control, Diabetes Mellitus control method, 0 or Oral or Insulin, Electrical Medical Record vii. HTN, Hypertension, 1 or 0, Electrical Medical Record viii. PAOD, Peripheral artery occlusion disease, 1 or 0, Electrical Medical Record ix. Dialysis, Dialysis, 0 or HD or PD, Electrical Medical Record x. OldCVA, Old Cerebral Vascular Accident, 1 or 0, Electrical Medical Record xi. CarotidStenosis, Carotid Artery Stenosis, 1 or 0, Electrical Medical Record xii. LiverDisease, Liver Disease, 1 or 0, Electrical Medical Record xiii. COPD, Chronic Obstructive Pulomnary Disease, 1 or 0, Electrical Medical Record xiv. Alcohol, Alcohol consumption, 1 or 0, Electrical Medical Record xv. Tobacco, Tobacco use, Never or Current or Former, Electrical Medical Record xvi. NYHA, NYHA Functional Class, 1, 2, 3, 4, Electrical Medical Record xvii. Afib, Atrial Fibrillation; paroxysmal, persistent, permanent, Electrical Medical Record xviii. LVEF, Left Ventricular Ejection Fraction on preoperative echocardiography(%); 1 - 100, Electrical Medical Record xix. Procedure, Surgical Procedure, "AVR MVR DVR CABG Aorta others", Electrical Medical Record xx. XCtime, Cross-Clamp time (mins), 0 - 1000, Electrical Medical Record xxi. CPBtime, Cardiopulmonary bypass time (mins), 0 - 1000, Electrical Medical Record xxii. delay.closure, delayed sternal closure, 1 or 0, Electrical Medical Record xxiii. SIMA, Single internal mammary artery graft, 1 or 0, Electrical Medical Record xxiv. BIMA, Bilateral internal mammary artery graft, 1 or 0, Electrical Medical Record xxv. BUN, Blood Urea Nitrogen , 0 - 100, Hospital Lab xxvi. Cr, Serum Creatinine , 0 - 20, Hospital Lab xxvii.AST, Serum Aspartate Aminotransferase level , 0 - 1000, Hospital Lab xxviii.ALT, Serum Alanine Aminotransferase level , 0 - 1000, Hospital Lab xxix. WBC, White blood cell count , 0 - 10, Hospital Lab xxx. Albumin, serumAlbumin , 0 - 10, Hospital Lab xxxi. Prealbumin, Serum Prealbumin level , 0 - 50, Hospital Lab xxxii.TRF, Serum Transferrin level , 0 - 200, Hospital Lab xxxiii.Serum Zinc level, Serum Zinc level , 70-120, Hospital Lab xxxiv.TG, serum triglycerol level , 0 - 1000, Hospital Lab xxxv.CHOL, serum CHOL level, 0 - 1000, Hospital Lab xxxvi.HDL, serum high density lipoprotein level , 0 - 100, Hospital Lab xxxvii.LDL, serum low density lipoprotein level , 0 - 1000, Hospital Lab xxxviii.HbA1c, hemoglobin A1c level , 0 - 100, Hospital Lab xxxix.Vanco.POD0, Serum Vancomycin level on postoperative day 0, 0 - 50, Study lab xl. Vanco.POD3, Serum Vancomycin level on postoperative day 3, 0 - 50, Study lab xli. Vanco.POD7, Serum Vancomycin level on postoperative day 7, 0 - 50, Study lab xlii. Stroke, postoperative Stroke, 1 or 0, Electrical Medical Record xliii. Bacteremia, postoperative Bacteremia, 0 or the bacteria species, Electrical Medical Record xliv. Pneumonia, postoperative Pneumonia, 1 or 0, Electrical Medical Record xlv. Resistance.pathogen, Resistance pathogen, 1 or 0, Electrical Medical Record xlvi. Intubation.days, Intubation days, 0 - 100, Electrical Medical Record xlvii. Mechanical.support.type, postoperative Mechanical support type, "IABP ECMO VAD", Electrical Medical Record xlviii. Parenteral.nutrition, postoperative Parenteral nutrition, 0 or PPN or TPN, Electrical Medical Record xlix. ICU.days, ICU days, 1-100, Electrical Medical Record l. Hospital.days, Hospital days, 1-100, Electrical Medical Record li. ThirtyDays.mortality, 30 days mortality, 1 or 0, Electrical Medical Record lii. Hospital.mortality, Hospital mortality, 1 or 0, Electrical Medical Record liii. Wound.POD7, Sternal wound condition on postoperative day 7, normal or superficial sternal infection or deep sternal infection, Study Nursing Practitioner liv. Wound.POD30, Sternal wound condition on postoperative day 30, normal or superficial sternal infection or deep sternal infection, Study Nursing Practitioner lv. Wound.POD90, Sternal wound condition on postoperative day 90, normal or superficial sternal infection or deep sternal infection, Study Nursing Practitioner 2. Data Collection: i. Clinical parameters collection. All patient demographic data and parameters of STS (society of thoracic surgery) database will be recorded, such as age, sex, cause of cardiac diseases, co-morbidities, concurrent medication, body mass index, nutrition status, operation details, days of hospital and ICU stay, hospital cost, any surgical complications, and any documented infection and their microbiology susceptibility test results. ii. Biochemistry and blood cell count. Data such as creatinine levels, fasting sugar levels, HbA1C, liver enzymes, nutrition status, and white cell count will be measured on postoperative day 0, 3, and 7 in the central laboratory of National Cheng Kung University Hospital. 3. Data management. i. Two registrars are responsible for data registry. Data are stored in spread sheet format. ii. Missing Data. All data source are available in NCKUH's electrical medical record. Any variable with missing data more than 3% will be dropped. 4. Report for adverse events. The investigators are responsible for adverse events reporting to our institution review board (NCKUH IRB). The reporting procedure is generally based on the form from the Council for International Organizations of Medical Sciences (CIOMS). The definition and requirement for Serious Adverse Event (SAE), Suspected Unexpected Serious Adverse Reaction (SUSAR), and Unanticipated Problems (UP), are available at our IRB website: http://nckuhirb.med.ncku.edu.tw/upload/download/file20191217230754.pdf. 5. Quality assurance. This is a single institution study. Quality assurance will be done by monthly investigator meeting in our hospital, which including the following: i. Data checks. Monthly data visualization report, either histogram for continuous variable or bar chart for categorial variable, will be used for detection of any data running out of range or having inconsistency with predefined rules. ii. Source data verification. Accuracy and completeness of the registry data will be double checked by the primary care nursing practitioners by comparison with the electrical medical record system. \------------------------------------------------------------------------------------------------------- 5\. Data Analysis, Statistical and Pharmacokinetic Analysis 1. For clinical data, continuous variables are expressed are as means ± standard deviation and analyzed with t-test. Categorical variables are presented as numbers (percentages) and Pearson's χ 2 or Fisher's exact test was used where appropriate. Univariate and multivariate logistic regression model and Cox model will be used to analyze risk factors for sternal wound infection. Significance was set at P \< 0.05. Statistical analysis was performed using R, version 3.2.5 (R Foundation for Statistical Computing, Vienna, Austria.). 2. For serum Vancomycin levels, non-compartmental analysis will be conducted with WinNonlin software (www.certara.com). The population pharmacokinetics will be analyzed with nonlinear mixed-effects modeling using in NONMEN 7.3 (Icon Development Solutions, Hanover, MD). Xpose (xpose.sourceforge.net), PsN (psn.sourceforge.net), and R, version 3.2.5 (R Foundation for Statistical Computing, Vienna, Austria.) were used for postprocessing of NONMEM output.