Sternal Closure With Biologic Bone Cement in Patients Undergoing Coronary Artery Bypass Grafting (CABG)

Procedure Research assistants will be dedicated to participant recruitment, data collection, and assessment of sternal wounds. Upon admission, or scheduling for surgery, consecutive CABG surgery patients who will have median sternotomy will be approached by a research assistant for preoperative screening (to determine eligibility for the study). Thereafter, informed consent will be sought to examine their recovery for a period of approximately 16 post-operative weeks.

When undertaking the clinical trial, randomization of the participants will be achieved using an opaque envelope. Envelopes will be kept at the operation theatre and will be opened by a nurse or the surgeon prior to surgery. This envelope will also contain a data collection form to be completed by the surgeons (regarding bone quality, sternotomy, and number of sternal wires).

Following surgery, blinded research assistants will collect pain data from the participants on the 1st to 5th postoperative days. These will be measured using a 10-point numeric rating scale. The scales will be anchored with descriptors 0=none, 10=worst. These data will be collected at rest and after coughing. Lung function data (FEV1.0, FVC) will be collected using a bedside spirometer on the 3rd and 5th postoperative days. Other relevant clinical data (e.g., age, grafts used for and nature of surgery, analgesic use, other complications, length of stay) will be collected through a health record audit.

Once the participants are discharged, data collection will be done when they return to the clinic for follow-ups.

Additional Outcomes Other outcomes that will be monitored will include: postoperative death within 6 weeks, abnormal prolongation in wound healing, length of ICU stay, length of hospital stay, postoperative pleural effusion, necessity to evacuate effusion, atelectasis, pneumonia, reoperation requiring sternal reentry, sternal dehiscence, wound infection (superficial or deep), mediastinitis, amount of analgesia, and all other causes of morbidity.

Sample Size There is not much information prior to this study. It has been assumed that the mean differences for the changes in FEV1.0, FVC and pains score at 5 days after surgery between bone cement and control groups to be 15% and the standard deviation to be 25% respectively. With these assumptions, a two-sided test, overall level of significance at 5% and power of 80%, the number of patients that is needed to be enrolled in this study is 64 patients per group after accounting that there will be 1% chance that the patient may have some major postoperative complication that will prevent the patient from being extubated at 5 days after surgery. The total number of patients that need to be recruited in this study is 128 patients.

Data Collection Participants will be recruited and in-hospital data collected by research assistants hired for this study. Staff nurses in the wards will also collect in-hospital data. They will be trained to use the study questionnaires and measures. Once patients are discharged from the hospital, the assistants will also follow-up in the outpatient clinic.

Data Analysis For this study, the investigators wish to assess the effectiveness of KryptoniteTM on pain, and change in postoperative lung function using spirometry.

Although the investigators are interested in the outcomes over time, the effectiveness of the intervention during the acute period postoperatively as well as 6 postoperative week will be of greatest interest. In order to investigate the difference in the percentage change in FEV1.0, FVC and pain score at 5 days after surgery between bone cement and control groups, 2 sample t-test will be carried out respectively if normality and homogeneity of variance assumptions are met; otherwise, the Mann-Whitney U test will be used respectively. The association between postoperative death in the early period and surgery type group (bone cement or control group) will be evaluated using Chi-Square test if the assumptions for the Chi-Square test are met; otherwise, Fisher's Exact test will be carried out. Other outcomes that are monitored will be summarized by the surgery type group using descriptive statistics. All the analyses will be performed using PASW Statistics 18 (SPSS Inc., Chicago, IL, USA).

SAFETY

Safety, particularly related to sternal union and wound healing, are of major importance in this study. In particular, sternal union will be monitored for all patients by visual inspection. All participants will undergo a visual inspection and palpation of their sternum, by the team doctors and nurses, regularly throughout the course of this study. Should any concerns arise with the sternum or wound healing, a digital photograph of the sternum (using appropriate draping for patient privacy) may be taken, and referred to his/her consultant surgeon in charge. The photographs will be used to assist in documenting the healing processes.

DESIGN ISSUES There are a number of important issues that are related to blinding, design and co-intervention.

Addressing Issues of Observer Blinding l All those collecting data will be trained rigorously in the standardized study procedures and use of the questionnaires and measures.

l The nature of some of the data to be collected is only the patient's opinion/response.

l A manual of operations will be developed for research associates' use and assistance with any question/concerns will be available by telephone at all times.

l Although double blinding is desirable in clinical trial designs to minimize/eliminate observation bias, it is not possible to blind the surgeon. The data collection members will be kept blinded regarding the grouping of the participant.

Addressing Issues of Participant Blinding l The outcome measures for this study are participative in nature, particularly as compared to those often used in clinical trials research (e.g., mortality/morbidity). Yet, these outcome data are highly relevant to determining the effectiveness of using a sternal closure technique such as this, in this particular population. As stated in the project objectives, it is one of our goals to evaluate the effect of the intervention on pain and discomfort; both of which are important to the patient.

l A 'placebo' is neither achievable nor feasible. Surgeons involved in participant's care will not divulge the group to which the participant was assigned.

ETHICAL/SAFETY CONSIDERATIONS

The conduct of this study will conform to the standards of DSRB. Data management will be undertaken using DSRB standards for maintaining patient confidentiality.

FEASIBILITY Cardiac surgeons at National University Hospital will offer the intervention for this study. The investigators anticipate it will take approximately 12 months to recruit the study sample and 4 months thereafter to complete data collection.