Prehabilitation in Liver Surgery
Introduction Thirty per cent of patients with colorectal cancer have metastatic disease at time of presentation, and a further 20% will develop liver metastases after the primary colorectal malignancy has been resected. Liver resection offers the prospect of cure for a proportion of these patients and, with the increasing use of effective neo-adjuvant chemotherapy, this proportion is increasing. However, liver surgery is associated with significant morbidity and mortality and this may be higher in patients with comorbidity, poor cardiorespiratory fitness and in those who have received neo-adjuvant chemotherapy.
Cardiopulmonary exercise testing (CPET) is a non-invasive assessment of cardiovascular and pulmonary function, which can be quantified by measures such as anaerobic threshold (AT) and VO2peak. The anaerobic threshold is a measure of sustainable exercise, where the VO2peak is a measure of maximal exercise capacity. The AT has been assessed in different surgical groups and has been shown as a useful predictor of postoperative complications and survival. Early work has demonstrated that short periods of preoperative exercise intervention can improve AT and VO2peak. However, no work has to date been undertaken in patients prior to liver surgery.
Hepatic glucose metabolism provides much of the energy requirements of the postoperative period. Work has demonstrated that exercise training increases hepatic glucose production, and that its inhibition has a marked effect on endurance capability. This may be particularly relevant when surgical resections can involve resection of up to 80% of hepatic tissue. However, a link between hepatic gluconeogenic capacity and fitness as assessed by cardiopulmonary exercise testing has not been established.
A demonstrable link between gluconeogenic capacity and cardiopulmonary fitness and an explanation of its underpinning physiology would help explain some of the systemic effects of drug hepatotoxicity. It would also allow development of strategies to improve gluconeogenic capacity that may reduce complications and improve tolerance of many hepatoxic agents such as chemotherapy.
Hypothesis
1. A short exercise program can significantly improve the CPET defined anaerobic threshold prior to liver surgery in a cancer population and this will be associated with reduced post-operative morbidity and mortality.
2. Hepatic gluconeogenic capacity will be associated with cardiopulmonary fitness.
3. Gluconeogenic capacity will be associated with mitochondrial number and quantity of enzymes which are integral to gluconeogenesis such as PEPCK
Methods Patients with colorectal liver metastases suitable for hepatic resection will undergo a baseline CPET and then be randomised to either preoperative exercise intervention or standard care. Patients within the intervention arm will undergo a 4 week exercise program consisting of 3 interval sessions per week on a stationary bike. This will be individually tailored according to their initial exercise test. In the week prior to surgery all patients will then undergo a further CPET. At surgery liver tissue will be taken to determine hepatic gluconeogenic capacity. CPET tests will then be performed where possible in patients at 6 weeks and 3 months following surgery. Post-operative complications (Clavien classification and Postoperative morbidity score (POMS)) and 30 and 90-day mortality will be recorded. Quality of life assessments (EORTC/SF-36) will be taken at recruitment, the week prior to surgery, 6 weeks after surgery and 3 months following surgery.
Laboratory Analysis Hepatic gluconeogenic capacity will be assessed by direct analysis of fresh slices of hepatic tissue taken at the time of surgery. Slices will be taken using the Krumdiek MD6000 tissue slicer. These will be weighed an incubated in a buffer containing lactate. Glucose will be measured using the Glucose Oxidase assay, and calculated per mg wet weight of hepatic tissue. Further analysis of mitochondrial number and ATP production, and enzymatic levels will be conducted following initial results.
Statistical considerations This is a study of a continuous response variable from matched pairs of study subjects and will need 38 patients to detect a true difference 1.5ml/min/kg, with a probability 0.8 and Type I error 0.05). Anticipated recruitment period is 10 months assuming a dropout rate of 20%. Randomisation is by computerised block randomisation, and patients will be stratified by receipt of neo-adjuvant chemotherapy.
