Feasibility Study of Early Review and Early Cardiac Rehabilitation After Coronary Artery Bypass (FARSTER-I)

Rationale and Background

Currently following CABG, patients attend their first outpatient review six weeks after hospital discharge, where recovery is assessed and fitness to commence CR is determined. CR is then started from eight weeks. In a survey conducted in May/June 2017 by the investigators, 35 of 42 UK cardiac centres responded, and confirmed this as current practice. The long interval before postoperative review and CR extends the period of vulnerability and inactivity for patients, with patients often seeking medical attention for surgery-related complications during this period. In a prospective observational study (FORCAST6), the investigators found that 44% of patients sought further medical help during this 6 week period, and 41% would like an earlier review. This feasibility study will examine the feasibility of bringing forward outpatient review and CR in order to facilitate recovery.

CABG is usually performed through a cut in the breastbone (median sternotomy) and patients are required to refrain from upper body exercises, lifting of heavy objects and usual strenuous activities for 12 weeks. These restrictions, called "sternal precautions" (SP), are intended to aid healing of the cut breastbone. CR which has significant short- and long-term benefits after CABG is therefore delayed. The delay can mitigate the benefits of CR, contribute to physical deconditioning, and hinder the ability of CR to facilitate timely recovery of fitness and physical activity status. According to the 2013 UK national audit of CR, late CR commencement contributes to a substantial number of heart surgery patients declining to participate in CR. Current time frames for postoperative review and CR are not evidence-based and may prolong recovery after CABG, and be placing unnecessary limitations on patient activity.

The standard access for heart operations is through a median sternotomy. In the UK, heart operations have steadily increased since 2010. Of 36,134 operations performed in 2013, 17,630 were isolated CABG. Following CABG, the recovery pathway after hospital discharge has two important elements, namely; outpatient review and CR.

First postoperative outpatient review: It is routine practice to undertake the first postoperative review six weeks after hospital discharge. However, high rates of hospital readmissions for surgery-related complications within this period have been reported. The investigators conducted a prospective observational study (FORCAST6) to investigate postoperative morbidity during, as well as patient opinion about, the six-week interval between hospital discharge and first review. This study showed that 44% of patients reported surgery-related complications; 17.5% required hospital readmission. Although majority of patients who did not have complication were satisfied with the six weeks interval, 41% felt it was too long. Early contact with patients has been associated with reduction in hospital readmissions. Non-randomised studies have also shown lower 30-day adverse outcomes with early review, for heart and other surgical patients.

Outpatient exercise-based cardiac rehabilitation: In current practice, CABG patients commence CR from eight weeks after hospital discharge. Current guidelines for activity and exercise after median sternotomy have been described as restrictive, conflicting, sometimes arbitrary, frequently anecdotal and occasionally based on expert opinion [6]. There are variations in types of exercises permitted, limits for weight of objects that can be lifted, guidelines for activities, and timeline for resumption of driving. Parker et al demonstrated that the force elicited on the breastbone by coughing far exceeded lifting above the recommended limit. Adams and colleagues investigated forces associated with 32 activities of daily living and reported that the majority not restricted by SP, such as opening and closing doors, generated forces greater than the weight limit allowed. While SP may help to support bone healing, the optimal nature and duration are unclear especially since sternal bone healing occurs by five weeks. A prior observational study (FORCAST6) by the investigators showed that the incidence of postoperative complications is highest in the first week after hospital discharge and declines to lowest levels by 4 weeks. CR can therefore be safely commenced 5 weeks after surgery (4 weeks after hospital discharge).

Early initiation and progressive physical exercise after heart surgery have well known benefits and prolonged restrictions of physical activities appear detrimental to patient health and wellbeing. LaPier et al reported functional limitation following sternotomy and related this to surgeon-dictated SP. A small observational study noted that functional recovery to pre-surgery levels was achievable at two months with physical therapy. Improvement in post sternotomy symptoms is also associated with physical activity. The timing of CR impacts functional recovery and level of physical fitness after heart surgery. The National Audit of CR 2015, recommends early commencement of CR which has become standard practice after coronary artery stents or heart attack.

A delay in starting CR can prolong the recovery process, increase dependence on family and/or carers and can cause frustration. This may contribute to anxiety and depression that is reported in patients recovering after CABG.

Published data is lacking about the appropriate timing of postoperative review and CR following CABG. The dominant role of SP as a deterrent to early CR after heart surgery has been questioned. Cahalin and associates identified major setbacks with current SP practice to include; lack of supporting evidence, variation between institutions, lack of universally accepted definition, and the functional limitations they engender. A similar experience was reported in Europe and Australia. These inconsistencies create a dilemma for patients, their families, and healthcare professionals alike. Nevertheless, SP continues to inform conventional practice and recommendations of The British Heart Foundation.

Study goals and objectives

This study aims to establish the feasibility of delivering and evaluating a definitive RCT of; current 8th week versus 4th week CR commencement pathways, reducing the period over which patients are required to refrain from usual physical activities.

Objectives

1. To determine the feasibility of delivering outpatient review three weeks after discharge post CABG, followed by CR from four weeks.

   To achieve this, the study will:

   • Collect qualitative data through interviews, diary entries and focus group meeting with key research and clinical staff.
   • Identify any organisational, operational and clinical barriers documented by research staff.
   • Identify additional costs or cost savings associated with the proposed pathway.

2. To test the feasibility of running a fully powered RCT comparing current with the proposed pathway.

   This will include:

   • Assessing surgeons' willingness to conduct outpatient review three weeks after discharge and refer patients to CR, from research staff records.
   • Examining barriers to patient enrolment using contemporary record of the consent process, by research staff.
   • Identifying recruitment rates and drop-out to follow-up.
   • Identifying the most appropriate outcome measures.
   • Testing follow-up procedures and data collection tools and management.
   • Planning the economic evaluation that would accompany the definitive RCT.
   • Gathering outcome data for power and sample size calculations for the RCT.
3. To inform any necessary redesign of a new recovery pathway in light of information gained from the above.

Research Questions

1. What are the barriers and facilitators to providing the proposed pathway, as these will be multi-faceted and context specific?
2. Will the proposed pathway lead to better physical fitness and quality of life after CABG, compared to current practice?
3. What would be the barriers to conducting a RCT to determine clinical and economic benefits of the proposed pathway?
4. What would be the key outcomes for a definitive RCT, and how would they best be measured?

Study design

This is a feasibility study designed as a Randomized Controlled Trial using mixed quantitative and qualitative methods.

It is a feasibility study where, early review is conducted three weeks after hospital discharge following coronary artery bypass grafting (CABG), and CR from 4 weeks. This will be a multi-centre RCT of 100 patients comparing the proposed pathway with conventional practice.

Due to lack of literature in this area; unknown physician and patient acceptability of the intervention and the intervention's influence on recruitment and retention, this trial has been designed as a mixed methods study.

Study sites

Patients will be recruited at Castle Hill Hospital (Hull and East Yorkshire Hospitals NHS Trust), the main study site, and James Cook University Hospital, Middlesbrough (South Tees Hospitals NHS Foundation Trust).

The York Trials Unit (YTU) at University of York will be responsible for project and data management for this study.

Study details

Enrolment: A Research Nurse will approach prospective patients who are admitted for CABG through a median sternotomy, and provide them with an information sheet explaining the research. The patients will be encouraged to read the information and make a decision about participation prior to hospital discharge. Study patients will undergo surgery as normal and receive standard in-hospital postoperative care.

From the 4th day after surgery, the medical staff will screen prospective participants for eligibility by review of the results of routine clinical examination including sternal stability, and postoperative day 4 tests, namely: electrocardiogram, chest x-ray, full blood count, blood biochemistry, and transthoracic echocardiography when clinically indicated. The Research Nurse will obtain informed consent from eligible and willing patients, administer EuroQol five dimensions (EQ-5D-5L) questionnaire and request cardiopulmonary testing (CPEX) for the first 50 patients that are recruited. Patients' baseline data including data from these tests will be collected at enrolment.

Allocation: Enrolled patients will undergo 1:1 randomisation to either; current (control arm) or the proposed pathway (intervention arm). Randomisation will be undertaken using a centralised randomisation service provided by a York Trials Unit statistician not involved in recruiting patients, and will be stratified by site using randomly permuted variable block sizes.

Study intervention: Control (current pathway) - postoperative outpatient review 6 weeks after hospital discharge, followed by commencement of CR from 8 weeks. Intervention (proposed pathway) - postoperative outpatient review 3 weeks after hospital discharge, followed by commencement of CR from 4 weeks. The structure of outpatient review and CR will be the same for both arms of the trial, as in current practice, specifically:

1. First postoperative outpatient review: Specialist surgical team will perform postoperative outpatient review, as is standard practice, for all study patients. Postoperative history, clinical examination including sternal stability, chest x-ray, electrocardiogram and medication review will be undertaken and, patients that are certified fit for CR would be referred to start CR programme. This review will take place at six weeks post hospital discharge in the control arm, and at three weeks in the intervention arm. Patients considered unfit for CR will be given a further 3 week review appointment and the reason(s) documented.
2. Outpatient exercise-based CR: Patients referred for CR will be offered a comprehensive programme. A first appointment would be made with a cardiac specialist nurse who would provide the patients with advice and leaflets on cardiac risk factor reduction. This will typically include information on medication, diet, exercise and physical activity as well as psychosocial wellbeing and smoking cessation. Patients will be invited to a group education session on cardiac risk factor reduction. This will be delivered by specialist cardiac rehabilitation staff. At this first appointment, referral to other healthcare professionals such as specialist counsellors, pharmacists and dieticians may also be considered when necessary. Unless contraindicated, referral to the exercise component of the CR programme will be made. Before joining the CR exercise class, patients will receive a holistic assessment from a specialist physiotherapist or exercise professional. This will involve exercise testing using the incremental shuttle walk test (ISWT). This baseline test will help to personalise exercise prescription for each patent. Following this assessment, patients will be enrolled in exercise programmes. CR exercise training will consist of supervised low-to-moderate intensity exercise performed once or twice week for eight weeks (as is standard practice). Exercise will be prescribed according to standards published by the British Association for Cardiac Prevention and Rehabilitation and, the Association of Chartered Physiotherapists in Cardiac Rehabilitation. Exercise training will be performed in a gym-like environment with other patients. Interval circuit training is the most commonly prescribed mode of exercise with each individual exercise programme tailored to patients' specific needs and fitness level. The following equipment will be used; heart rate monitors, treadmill, static bikes, and hand weights. At the end of CR, a reassessment, including exercise testing using ISWT will be conducted. The pre and post CR tests will be recorded. A discharge letter would be sent to patients' General Practitioners summarising their treatment.

The control arm will commence CR at 8 weeks while the intervention arm will start at 4 weeks.

All study patients and CR staff will be encouraged to complete bespoke CR diaries for the duration of CR. Patients will complete EQ-5D-5L questionnaires at the end of CR.

Sample size

The research is a feasibility study and therefore does not have a primary outcome measure to inform a power calculation. One of this study's objectives is to establish a primary outcome and its variability to inform sample size calculations for subsequent research. Sample sizes of between 24 and 70 have been recommended for feasibility trials to allow for the reliable estimation of a standard deviation for use in future sample size calculations. The investigators hope to recruit 100 eligible patients. This will allow for a 30% attrition rate to still have 70 patients in the final analysis. There will be a 5 month recruitment period across two centres. The two clinical sites each perform over 300 isolated CABG operations annually, and it is anticipated that about 200 of these per site will be eligible for the study (100 per site over 5 months). The investigators therefore require a participation rate of 33% to meet our target sample size. If 300 eligible participants are identified, the investigators will be able to estimate a participation rate of 33% to within a 95% confidence interval of ±6%.

Quantitative tests and data collection

In this study ISWT and CPEX would be used to assess physical fitness. The ISWT is a field test and fairly crude measure of aerobic fitness. It is a popular aerobic fitness assessment tool in UK CR programmes. For this reason it is important that we use an ecologically valid tool to assess aerobic fitness if the findings from our study are to be extrapolated to routine clinical practice.

CPEX is not just a tool to assess aerobic fitness; it is the gold-standard method of assessing aerobic fitness and other prognostically important cardiopulmonary variables. CPEX generates far more information on the cardiovascular and cardiorespiratory health of patients meaning that we can characterise disease severity in a more meaningful way. CPEX produces a lot of data that may help explain mechanisms for change which ISWT is not able to yield unless we were to use telemetry base VO2 using something like COSMED. CPEX is the most accurate method of assessing fitness change. This is potentially important if we are to assess the beneficial/not harmful effects of early CR. Furthermore, the ISWT is not yet validated for use in the early CR cohorts.

It is important and worthwhile to collect some data on CPEX in this trial, however, given the cost of CPEX this will only be used in the first 25 participants randomised to each group, i.e. half our intended sample size.

The investigators would use the field test (ISWT) for all participants and also assess half the intended sample size with CPEX. This will reduce the costs but still give use useful information to (1) understand any physiological change and (2) validate ISWT against CPEX at baseline testing. This will also help to understand if they are measuring similar attributes of fitness.

Quantitative data will be collected for all the study patients from enrolment to completion of follow-up, using:

• Case report forms to collect general health and EQ-5D-5L questionnaires at baseline (prior to randomisation), completion of CR and 26 weeks after hospital discharge,
• Medical records of outpatient clinical review and exit consultation, listing hospital readmissions and, attendance at General Practitioners' surgeries and Accident and Emergency departments for surgery-related complications.
• CR report of exercise testing results (ISWT) at start and end of CR, and at 26 weeks post discharge,
• CPEX reports at baseline and 26 weeks post discharge The study design provides some challenges for collection of comparable data because of the difference in the timings of the components of the recovery pathway for the two study arms. For example, data collected at outpatient review, beginning and end of CR will be at different times in the recovery pathway for the 2 study arms. However changes in the parameters within each study arm would yield useful information about the trend. Having considered these challenges, the investigators intend to collect similar data for both study arms at baseline prior to randomisation and at 26 weeks follow-up, for direct comparison, even though late follow-up may carry a risk of high drop-out.

All data will be anonymised; patients will be identifiable only by a unique trial ID. Data collection will be monitored and inconsistencies, validation errors or inaccuracies will be followed up.

Qualitative data recording

Qualitative data will be collected from patients and, the research and clinical staff through:

• Face-to-face semi-structured interviews with patients at both the outpatient review and follow-up review at 26 weeks. Ten patients in each arm would be interviewed twice, ie 40 interviews. The topic guides for the interviews will use open ended questions to guide respondents to consider their experiences and views of the timing of the outpatient review, the offer of CR made to them, and, for the end of study interviews, their experiences of attending CR. Interviews will be digitally recorded and transcribed verbatim.
• Diary entries of patients, Research Nurse and CR staff. Patients would be given a bespoke diary that will cover the 8-week period of CR, with tick boxes and free text sections. They would be encouraged to make entries on a weekly basis during the 2 months of CR. This will capture postoperative complications, their experiences of pain and any health-related adverse occurrences during exercise-based CR. The Research Nurse shall keep a diary with a record of patients' and surgeons' attitude to recruitment for the study.

CR staff would also keep one diary at each of the 2 centres and will collect information about patient drop-outs or opt-outs of CR sessions, level of participation and adverse events during the sessions.

• A focus group meeting of the research and clinical staff involved in the trial would be held after the 26 week follow-up period. The focus group topic guide will use open-ended prompts to facilitate discussion on both individual and shared views of the CR offer made to patients in each treatment arm, and the barriers and facilitators at individual, clinical and organisational level associated with each treatment trajectory.

Safety consideration

Study patients will undergo clinical assessment after discharge following CABG, and be certified fit for CR by the surgical team. Before commencing exercise-based CR they will undergo exercise testing, and CR will be tailored to their fitness levels. At completion of CR, they would have another assessment and exit consultation. This would enable early detection of possible complications such as:

Sternal pain: Delayed sternal healing by forces or pressure not severe enough to disrupt healing contributes to severe pain associated with tenderness along the sternotomy. This may be confused with muscular or sternal wire pain, and paraesthesia which are more common.

Sternal instability: This is abnormal movement between the surgically divided sternum due to delayed healing or non-union. This complication is rare.

Any study patient who complains of sternal wound pain or a feeling of "sternal click" will be seen urgently by the surgical staff, examined and investigated appropriately. All adverse events will be recorded. Patients who suffer adverse events with causality linked to sternal instability will be given appropriate treatment, including surgery if required.

Follow-up

At 6 months (26 weeks) after surgery, all study patients will complete EQ-5D-5L quality of life questionnaire, have CPEX, undergo ISWT, and have an end-of-study consultation with the research staff, who will inquire about details of further medical treatment after hospital discharge. Patients will be examined for sternal wound complications, including sternal instability.

Statistical analysis

Statistical analysis will be undertaken in Stata version 13, or later, by a medical statistician at York Trials Unit. The analysis and reporting of this feasibility trial will follow CONSORT guidelines and a flow diagram will depict the progression of participants through the trial. The number of patients screened, eligible, consenting and randomised will be summarised, with reasons for non-participation given where possible. Baseline data will be summarised by randomised group. No formal statistical comparisons will be undertaken. Continuous variables will be summarised using descriptive statistics (n, mean, standard deviation, median, minimum and maximum) while categorical data will be reported as counts and percentages. Participant outcomes will be summarised descriptively by randomised group and time point, including extent of missing data. As this is a feasibility trial, formal hypothesis for effectiveness will not be carried out.

The number of participants attending their review appointment and entering CR, and the time between surgery and these events, will be summarised by treatment group. Questionnaire return rates will be presented. Adverse events will be summarised descriptively.

Qualitative data analysis

Data will be analysed separately per research grouping ie patients, research staff and clinical staff. Data will be analysed using conventional content analysis, with the aim of detailing individuals' views regarding the timing of outpatient review, commencement of CR, and experiences of attending and undertaking CR.

Transcribed data from the interviews and focus groups will be downloaded into the qualitative data analysis computer software package NVivo, coded and analysed inductively using thematic content analysis.

Interviews: As interviews are to be conducted at two time points, the analysis will compare themes both within and across time periods to determine if patients' views and experiences differed over time. As patients will be purposively sampled by age, gender and ethnicity, the analysis will also aim to identify intersections between these groupings. The findings from the patient generated qualitative data will inform the feasibility and acceptability of the offer and timing of CR.

Patient Dairies: The free text data detailing patient's experiences of receiving CR over the eight week study period will be subject to thematic content analysis, where the richness of the data allows. If data are quite 'thin', an analysis that counts recurring items within the text will be used. These data will be triangulated with the themes arising from the patient interviews as confirmation or refutation of the findings from those data.

Focus Groups: The focus group analysis will pay particular attention to the interactions between group members and staff groupings in the development of themes, in order to determine the barriers and facilitators perceived by staff groups at individual, clinical and organisational level within each treatment trajectory.

Finally, the themes from the patient interviews staff focus groups will be compared to determine the similarities and differences between the themes in order to reflect a complete picture of the feasibility and acceptability of the offer of early versus late CR.

In combination these qualitative analyses will be used to inform the development of the logistical processes involved in delivering a large-scale multi-centre RCT.

Economic analysis

Health economic analysis will be provided by a University of York (Centre for Health Economics) health economist. Due to the feasibility nature of the proposed trial the analysis will focus on the consideration of questions related to the feasibility of conducting a full economic evaluation. Estimates of patient benefit, determined from the EQ-5D-5L will be summarised for the two arms of the trial, including the rate of response of relevant questionnaires. Exploration of the short and long term NHS resource use related to both interventions will be conducted, and relevant sources of informative data required for a full scale RCT identified.

Mortality and adverse event rates will also be monitored to determine the importance of long term extrapolation of related outcomes for any future analysis of a full scale RCT. As this is a feasibility trial, formal estimation of the cost-effectiveness of the respective interventions will not be carried out.