Evolution in perioperative management of patients undergoing oesophagectomy

Outcomes associated with oesophageal resection are subject to increasing scrutiny within the medical literature and, recently, by groups, such as the Leapfrog Group (http://www.leapfroggroup.org) in the USA, that represent the consumer. Many reports address individual issues, such as surgical volume and operative approach. Although these issues undoubtedly affect morbidity and mortality, they also reflect a degree of egotism on the part of surgeons, many of whom seem to believe that surgical judgement and technique are, if not the sole, at least the major variables influencing outcome. My first oesophageal resection at my current institution in 1991 involved an Ivor Lewis resection in a 60-year-old man with a T3 N1 adenocarcinoma who had had neoadjuvant chemoradiotherapy. The operating time was 6 h, blood loss 350 ml, and intraoperative fluids and urine output were respectively 11·5 litres and 300 ml. The patient remained intubated overnight, had patient-controlled analgesia, spent 2 days in intensive care and was walking around by day 4; he left hospital on day 13. Last week I carried out an Ivor Lewis resection in a 73-year-old man with a T3 N1 adenocarcinoma following neoadjuvant chemoradiotherapy. The operation took 5·5 h, blood loss was 200 ml, intraoperative fluids and urine output were respectively 2·5 litres and 150 ml. The patient was extubated in the operating room; pain control was by patient-controlled thoracic epidural. He spent 12 h in intensive care, walked before leaving the unit and was discharged on day 7. The patient, preoperative and surgical issues are similar in these two scenarios—the major changes over time have occurred elsewhere within the multidisciplinary team, specifically in anaesthesiology, critical care, pain service, physiotherapy and nursing. Despite the changes, many doctors still regard oesophagectomy as a lengthy operation with great blood loss, frequent major complications, unacceptable risk of death and poor quality of life for survivors. This has led some to advocate potentially less successful endoscopic or non-surgical oncological therapies, and to avoid surgery. However, recent publications from high-volume centres have demonstrated that mortality rates can, and should, be below 2 per cent. The author's department has performed a consecutive series of 174 resections over 6 years, with outcomes showing that under the best circumstances the mortality rate can approach zero1; this indicates that death should no longer be the major issue highlighted in national assessments. The starting point for changing perceptions and standardizing approaches to perioperative management lies in improving communication within the multidisciplinary team, beginning with the anaesthetists. Anaesthetists are a data-driven group of physicians, whose contact with patients is typically short-term (compared with that of surgeons) and often involves treatment algorithms, especially with regard to intraoperative fluid administration. Many of these algorithms were developed in an earlier era when volatile anaesthetics were nephrotoxic. Furthermore, algorithms for intraoperative fluid administration based on calculations of insensible losses and urine output are still in current anaesthesia textbooks. This has resulted in patients receiving large intraoperative fluid volumes that are increasingly recognized as causing detrimental cardiovascular, pulmonary, thromboembolic and renal effects2. Priano and colleagues3 have demonstrated that, under controlled circumstances, relative intraoperative oliguria need not be of clinical significance. For these ideas to become acceptable to our anaesthesia colleagues, surgeons must demonstrate that the operation can be accomplished without unexpected episodes of haemodynamic instability secondary to prolonged mediastinal manipulation or blood loss. Surgeons should be able to vary the operative approach according to patient physiology as well as tumour characteristics and location. While adhering to the principles of cancer treatment, they must at the same time communicate directly with the anaesthetist to coordinate the response to transient and expected episodes of hypotension. Several recent publications have suggested that ‘fluid restriction’ reduces complication rates, length of hospital stay and cost4,5, and that postoperative ‘fluid overload’ may be associated with poor outcome and increased mortality6. Surgeons and anaesthetists should now agree that standard fluid replacement algorithms are unnecessary5 and recognize that Brandstrup7 is correct when she states that ‘current standard fluid therapy is not at all evidence-based’. Of equal importance are the immediate postoperative issues of extubation, pain management and mobilization. In many centres surgeons and anaesthetists seem to have arrived at a consensus that after an oesophagectomy patients are best served by being ventilated overnight. Quite frankly, much of this comes from the fact that, after a long operation with significant blood loss and fluid administration, a ventilated patient is easier to manage. The alternative approach of extubation before leaving the operating room requires meticulous attention to immediate pain control and pulmonary toilet, but it also provides a faster recovery, fewer complications and earlier mobilization. There is little doubt that regional anaesthesia and analgesia techniques, specifically patient-controlled systemic or epidural analgesia, improve pain control and lead to earlier extubation and decreased pulmonary complications8,9. Minimally invasive and certain open approaches to oesophageal resection have been designed specifically to avoid painful incisions (particularly thoracotomy) and so minimize postoperative discomfort. Irrespective of the approach, good analgesia supervised by a pain service facilitates aggressive chest physiotherapy, early mobilization and, ultimately, early discharge from hospital. Surgeons and anaesthetists should have a mutual agreement that regional anaesthesia and analgesia are today a standard component of oesophagectomy management. This multimodal approach does not only improve outcome8,9, it is also the foundation for ‘fast-tracking’ in the recovery period, including initiating early mobilization, nutrition and discharge. ‘Fast-tracking’ is generally achieved by the application of standardized clinical care pathways. In the author's department the perioperative management plan involves such a care pathway that has evolved over many years. It deals with preoperative investigations, intraoperative management, postoperative pain control and nutrition, and has a goal of mobilizing the patient on the first day after surgery. This early mobilization proved to be one of the most difficult pathway items to initiate. Progress was made when patients and families were told that a smooth recovery should include being up and walking on the first postoperative day, and a further advance occurred when transfer out of the intensive care unit was made contingent on walking. Achievement of the goal is now routinely expected by all members of the multidisciplinary team and is reached in 95 per cent of patients. The above issues of perioperative management are a few of the many factors that affect outcome after an oesophagectomy. Zehr and coll- eagues10 have demonstrated that standardized clinical care pathways can affect outcome after major thoracic surgery. Such pathways are particularly important after oesophagectomy, but they will not be successful unless all members of the team feel that they are part of the process and are kept updated on results. Strong surgical leadership is the key. Continued success in improving morbidity and mortality rates with improved perioperative management pathways will allow the emphasis to shift to other more appropriate targets, such as cancer survival rates and quality of life.