A review of enhanced recovery for thoracic anaesthesia and surgery


Correspondence to: A. A. Klein Email: andrew.klein@papworth.nhs.uk


During the past decade, there has been a dramatic increase in the number of thoracic surgical procedures carried out in the UK. The current financial climate dictates that more efficient use of resources is necessary to meet escalating demands on healthcare. One potential means to achieve this is through the introduction of enhanced recovery protocols, designed to produce productivity savings by driving reduction in length of stay. These have been promoted by government bodies in a number of surgical specialties, including colorectal, gynaecological and orthopaedic surgery. This review focuses on aspects of peri-operative care that might be incorporated into such a programme for thoracic anaesthesia, for which an enhanced recovery programme has not yet been introduced in the UK, and a review of the literature specific to this area of practice has not been published before. We performed a comprehensive search for published work relating to the peri-operative management and optimisation of patients undergoing thoracic surgery, and divided these into appropriate areas of practice. We have reviewed the specific interventions that may be included in an enhanced recovery programme, including: pre-optimisation; minimising fasting time; thrombo-embolic prophylaxis; choice of anaesthetic and analgesic technique and surgical approach; postoperative rehabilitation; and chest drain management. Using the currently available evidence, the design and implementation of an enhanced recovery programme based on this review in selected patients as a package of care may reduce morbidity and length of hospital stay, thus maximising utilisation of available resources.

Thoracic anaesthesia and surgery is an expanding specialty due to the increasing prevalence of lung cancer and the development of new surgical procedures. Mortality from lung cancer remains high; one person in the UK dies as a result every 15 min [1]. In the majority of instances, surgical resection remains the best hope for cure [2]. Over the last decade, the number of pulmonary resections performed for lung cancer has risen from 3112 in 2001–2002 to 5265 in 2009–2010, an increase of almost 60% [3]. Figures from the Department of Health show that surgery for all forms of cancer now accounts for nearly a quarter of the national cancer budget [4]; if projections for a sustained increase in demand for cancer services are to be met, more efficient utilisation of resources will be needed in the current era of financial austerity. Minimising length of hospital admission has been identified as a key means to increase capacity whilst reducing costs, and is a major objective of the Cancer Reform Strategy [4]. A growing body of evidence suggests that inpatient stay after elective surgery can be reduced by the introduction of an enhanced recovery programme (ERP) [5], which is a series of evidence-based practices, serving to optimise the patient before surgery, minimise the physical and psychological stress associated with the procedure and promote restoration of function [6]. Numerous studies demonstrate the value of enhanced recovery programmes in the surgical management of colorectal [7], breast [8], pancreatic [9] and urological [10] malignancies; however, there is a relative paucity of work in thoracic surgery in general and lung cancer in particular. The National Institute for Health and Clinical Excellence (NICE) has recently recommended that further work should be undertaken into the benefits of an ERP in this group of patients [11]. This article reviews the evidence for individual peri-operative interventions in patients presenting for thoracic surgery, and may be used to form the basis of local enhanced recovery programmes.


A comprehensive literature search was performed on both MEDLINE and Embase using the National Health Service health database advanced search interface, which includes the Cumulative Index to Nursing and Allied Health Plus (CINAHL), the British Nursing Index (BNI) and the Allied and Complementary Medicine Database (AMED), accessing articles from 2000 to 2012. To achieve maximum sensitivity of the search and indentify relevant studies, the following thesaurus terms were applied: ‘pulmonary surgical procedures’; ‘early ambulation’; ‘thorax surgery’; and ‘perioperative period’. Free-text items searched for were: ‘enhanced recovery’; ‘patient recovery’; ‘fasting’; ‘risk factors’; ‘medical optimisation’; ‘day of surgery admission’; ‘length of stay’; ‘suction’; ‘early ambulation’; ‘mobilisation’; ‘discharge’; ‘quality indicators’; ‘healthcare’; ‘pre-admission’; and ‘pain relief’. The search results were limited to English language studies. A total of 813 studies were initially identified, and two authors (NJ and AK) selected 95 articles for more detailed study based on their relevance to the topic of enhanced and improved recovery after thoracic surgery. Each article was examined and recommendations were made based on the best level of evidence for each element [12].

Pre-optimisation of patients before surgery

It is a basic tenet of enhanced recovery that patients should be in the best possible condition for surgery, modifiable risk factors should be addressed and comorbidities optimised before surgery. Ideally, this is undertaken by the general practitioner in the period following diagnosis whilst awaiting treatment, or, at the latest, at pre-operative assessment. Table 1 summarises the key aspects of pre-optimisation included in this review.

Table 1. Recommendations for pre-operative care as part of an enhanced recovery programme after thoracic surgery
Area of practiceParameterAction
  1. DVT, deep vein thrombosis

Pre-optimisationAnaemiaInvestigated and treated before surgery if possible [13-19]
NutritionScreening for malnutrition and nutritional support given to risk patients [20-23]
SmokingAdvised to stop smoking before surgery and given appropriate support [23-27]
Medical therapyMedical therapy should be optimised before surgery [28-30]
PhysiotherapyPulmonary rehabilitation considered before surgery to improve exercise capacity [31, 32]
Pre-assessmentPre-operative clinicDetailed assessment carried out to facilitate same-day admission and reduce unnecessary cancellation [33-35]
Risk calculationUsed to facilitate informed consent and appropriate resource allocation [36, 37]
EducationPatients and their families should receive detailed oral and written information about hospital stay, the recovery process and discharge [38-40]
AdmissionSame-day admissionReleases bed capacity and minimises hospital stay [41]
PremedicationAnxiolytic premedication can be given without delaying recovery, but use with extreme caution [42]
FastingMinimise ‘nil-by-mouth’ and consider carbohydrate beverage 2 h preoperatively [43, 44]
DVT prophylaxisAll patients should be assessed and anti-embolism stockings or mechanical devices started [45]

Anaemia is a common incidental pre-operative finding and is associated with increased morbidity and mortality in the peri-operative setting [13]. It should be investigated and treated before surgery if time permits. Peri-operative blood transfusion is the commonest method of raising haemoglobin concentration in anaemic surgical patients, but it is associated with the risk of acute transfusion reactions, immunosuppression, postoperative infection and longer hospital stay [14, 15]. Alternative strategies to correct even minor degrees of anaemia (haemoglobin concentration < 12 g.l−1 in females, < 13 g.l−1 in males) before surgery, which have been noted to be highly prevalent in recent studies [16], can significantly reduce the need for transfusion and the resultant increase in morbidity and mortality [17]. Although studies have examined the utility of iron supplementation and erythropoietin in the context of patients with lung cancer undergoing adjuvant treatment with chemo- or radiotherapy, none have investigated their use before surgery [18, 19].

Malnutrition is another common finding in patients with cancer and is associated with impaired wound healing, immune dysfunction, respiratory muscle fatigue and tissue wasting following surgery. This results in delayed recovery and prolonged length of stay [20]. The European Society for Nutrition and Metabolism Guidelines [21] recommend that all patients should be screened for malnutrition, and those with severe increased risk (weight loss 10–15% within 6 months, BMI < 18.5 kg.m−2, Subjective Global Assessment Grade C, or serum albumin < 30 g.l−1) should receive nutritional support for 10–14 days before major surgery. Up to 28% of patients with operable lung cancer are reported to be at severe nutritional risk [22]. To date, no studies have specifically examined the impact of pre-operative correction of malnutrition in patients with lung cancer. One small prospective randomised trial has investigated the effect of micronutrient supplementation in patients with non-small cell lung cancer and a normal body mass index for 10 days before lung resection [23]. The combination of α-ketoglutaric acid and 5-hydroxymethylfurfural not only improved exercise capacity and reduced oxidative stress but also resulted in a significant reduction in intensive care and hospital stay. These findings require confirmation in larger studies

Smokers are more likely to die following surgery and have significantly greater odds of experiencing serious postoperative complications [24]. In patients with lung cancer undergoing pulmonary resection, smoking cessation reduces postoperative mortality and morbidity [25]. No evidence of a paradoxical increase in pulmonary complications among those who quit smoking within two months of undergoing thoracotomy has been demonstrated [26]. The NICE guidelines on the diagnosis and treatment of lung cancer [11] recommend that all patients should be advised to stop smoking as soon as the diagnosis of lung cancer is suspected; however, surgery should not be postponed to allow this. All patients should be offered nicotine replacement and other therapies to help stop smoking [27].

Chronic obstructive pulmonary disease (COPD) is very common in patients with lung cancer, and is a risk factor for peri-operative pulmonary complications. Whilst intuitively one would expect that pharmacological treatment would improve outcome, there are few data in patients with either newly diagnosed or established COPD undergoing lung resection to substantiate this. Data that exist suggest that commencing a long-acting bronchodilator in patients with untreated COPD significantly improves respiratory symptoms and pulmonary function [28]. The addition of inhaled steroid may also reduce postoperative complications [29]. Intensive pre-operative respiratory physiotherapy combined with optimised drug treatment (inhaled bronchodilators or corticosteroids) in patients considered unfit for surgical resection can result in significant improvements in lung function to the extent that they can be reconsidered as surgical candidates [30]. Pre-operative pulmonary rehabilitation may also be employed for selected patients with COPD and lung cancer to improve exercise capacity [31, 32].


All patients undergoing elective surgery should undergo pre-operative assessment [33] to establish that the patient is fit for the proposed surgery and anaesthetic. There is some evidence that pre-operative assessment may facilitate same-day admission, reduce unnecessary cancellations and increase patient satisfaction [34]. Assessments of exercise capacity such as shuttle walk tests and stair climbing, together with formal measurements of cardiopulmonary function, may help decision-making in patients at high risk of postoperative dyspnoea [35].

Accurate prediction of peri-operative risk is fundamental to the process of informed consent. It also serves to identify those patients at higher risk, who may derive greater benefit from pre-optimisation and require more intensive postoperative care. The most widely used clinical tool for predicting in-hospital mortality after general thoracic surgery is the Thoracic Surgery Scoring System (Thoracoscore). This system consists of nine variables, with a correlation between observed and expected mortality of 0.99 [36]. Other well-recognised risk factors for postoperative morbidity and prolonged length of stay include age, obesity, poor pre-operative lung function (FEV1 or TLCO < 40%), impaired functional capacity (VO2 max < 15 ml.kg−1.min−1), ASA physical status, continued smoking, insulin-dependent diabetes, chronic renal failure and regular pre-operative analgesic use [37]. Patients with multiple risk factors may require more intensive use of resources, such as high dependency postoperative care or prolonged hospital stay, and may need to be managed outside of the enhanced recovery programme. It is important that such patients are identified pre-operatively and appropriate arrangements are made. At the same time, there is evidence that exploring patients' expectations in the pre-assessment clinic may improve outcome [38, 39]. Detailed explanation of the intended peri-operative pathway has also been shown to reduce patient anxiety, pain and length of hospitalisation [40].

Admission to hospital

Effective pre-assessment means that hospitalisation for a day before surgery is unnecessary (Table 1). Same-day admission for patients undergoing surgery for lung cancer is practised in a number of institutions within the UK [41]. The administration of routine anxiolytic premedication has declined in recent years because of concerns regarding delayed recovery. However, a Cochrane review of 17 randomised clinical trials (RCTs) comparing anxiolytic premedication (benzodiazepines, opioids and beta-blockers) with placebo found no impediment to recovery [42]. Unfortunately, this review did not include patients undergoing thoracic surgery, many of whom have impaired respiratory function and in whom the use of sedatives could be hazardous. Thus, premedication can be employed in selected patients, but should be used with caution and increased vigilance. Same-day admission may lead to the use of premedication becoming rarer, and many institutions may choose to omit it altogether from their enhanced recovery programme.

Table 2. Recommendations for intra-operative care as part of an enhanced recovery programme after thoracic surgery
Area of practiceParameterAction
AnaesthesiaAntibiotic prophylaxisIn accordance with local policies and with knowledge of the patient's colonisation and resistance patterns [46, 47]
Choice of agentsUse short-acting agents that facilitate early recovery; inhalational anaesthesia may have advantages over intravenous techniques [49-54]
VentilationLungs should be ventilated using a protective strategy with limited tidal volumes [55-60]
FluidsAvoid fluid overload [61-63]
Atrial fibrillationProphylaxis should be considered in at-risk patients [64-68]
Tracheal extubationPatients' tracheas should be extubated at the end of surgery if possible [73]
SurgeryApproachMinimally invasive, provided it does not compromise the curative intent of the procedure [69, 70]
DrainsOne chest drain should be used in preference to two [71]
SealantsCannot be recommended [72]
AnalgesiaTechniqueParavertebral analgesia should be used in preference to thoracic epidural [74]

Pre-operative fasting causes metabolic and psychological stress. ‘Fasting from midnight’ has been common practice in the past to reduce the risk of pulmonary aspiration during anaesthesia and the immediate postoperative period. However, a review of the evidence demonstrates that the pre-operative fasting period for clear fluids can be reduced to 2 h without increasing complications [43], although specific data from patients of ASA status 3–4, such as many thoracic surgical candidates, are lacking. There is some evidence that carbohydrate loading is associated with faster recovery and shorter hospital stay [44].

Thoracic surgery patients should be regarded as high risk for postoperative venous thromboembolism (VTE). Recent NICE guidelines recommend that mechanical VTE prophylaxis (anti-embolism stockings, intermittent pneumatic compression devices or foot impulse devices) should be commenced at admission [45]. Pharmacological VTE prophylaxis should be added in patients who have a low risk of major bleeding, with either low molecular weight heparin or unfractionated heparin for patients with renal failure. Prophylaxis should be continued until the patient no longer has significantly reduced mobility. Care must be taken in the timing of administration of pharmacological prophylaxis in patients who may receive regional anaesthetic blocks for their surgical procedure.

Intra-operative management

There are a number of areas of intra-operative practice that may be considered for inclusion in an enhanced recovery programme (Table 2). Prophylactic antibiotics have been shown to reduce infectious complications after thoracic surgery [46]. However, the current evidence on what constitutes the most appropriate antibiotic prophylaxis for this group of patients is poor, and no guidelines exist. Importantly, these patients will often have suffered multiple episodes of respiratory tract infection, and received numerous courses of antibiotics before their surgery. Airway colonisation with pathogens is a risk factor for the development of postoperative pulmonary infectious complications [47]. When selecting suitable prophylaxis, changes in the usual pattern of flora as a result of chronic colonisation and the potential development of antibiotic resistance should be considered. Antibiotics should be given 60 min or less before ‘knife to skin’ in accordance with the World Health Organization (WHO) Safer Surgery checklist [48].

The use of short-acting anaesthetic agents seems rational to facilitate early recovery. Inhalational anaesthesia has the theoretical disadvantage of inhibiting hypoxic pulmonary vasoconstriction; however, in clinically relevant concentrations, it has no significant effect on shunt fraction [49, 50] or arterial oxygen content during one-lung ventilation [51]. One randomised study found that cerebral oxygen balance during lung surgery is less impaired under sevoflurane-based anaesthesia compared with propofol; however, the clinical implications of this finding need to be determined [52]. Various reports have cited the relative benefits of sevoflurane [53] or propofol [54] on peri-operative cytokine balance in patients undergoing one-lung ventilation, but the results are conflicting and no firm recommendations on choice of agent or technique can be given.

One-lung ventilation induces cytokine release and activation of the pulmonary inflammatory response [55, 56]. Use of large tidal volumes can exacerbate lung injury and is associated with an increased risk of respiratory failure after pneumonectomy. Protective mechanical ventilation, using low tidal volumes (5 ml.kg−1), has been shown to reduce serum inflammatory markers and the incidence of postoperative pulmonary dysfunction, such as relative hypoxia or newly developed lung infiltrates or atelectasis [57]. The effect of positive end-expiratory pressure on postoperative outcomes remains unclear [58]. Limited data suggest that pressure-controlled ventilation may be associated with improved intra-operative oxygenation and reduced mean airway pressure [59] compared with volume control. However, a recent review concluded that the evidence for an effect on postoperative oxygenation was inadequate [60].

Sodium and fluid overload are associated with increased postoperative complications and prolonged hospitalisation. In patients undergoing lung surgery, intra-operative fluids are frequently restricted as positive fluid balance is one of the strongest risk factors for the development of postresection acute lung injury [61]. In addition, or as an alternative, to conventional measurement of central venous pressure, several devices have been shown to predict fluid responsiveness in mechanically ventilated patients. In the main, these devices rely on measurement of the cyclical variations in stroke volume or pulse pressure induced by mechanical ventilation, but conditions during thoracic surgery (such as open chest, shunt during one-lung ventilation) may limit their usefulness. In practice, it has been shown that pulse pressure variation and stroke volume variation may predict fluid responsiveness under certain circumstances [62, 63], but further research specific to thoracic surgery is needed before widespread use for goal-directed fluid therapy can be recommended.

Atrial fibrillation occurs relatively commonly after pulmonary resection (12–30% after lobectomy), and is associated with significant morbidity and increased length of stay and hospital costs [64]. An analysis of the Society of Thoracic Surgeons database has identified five variables that predict postoperative atrial fibrillation: advancing age; increasing extent of operation; male sex; non-black race; and more advanced or large tumours [65]. No standard regimen has been recommended to decrease the incidence of atrial fibrillation; however, evidence from prospective RCTs supports the use of beta-blockers, diltiazem, intravenous magnesium or amiodarone for the prevention of atrial fibrillation after pulmonary resection [66, 67]. The use of amiodarone in this setting is controversial due to the risk of acute lung injury; however, studies attest to the safety of this drug if the cumulative dose is limited [68].

Surgical approach

The traditional approach to pulmonary resection is a postero-lateral thoracotomy as this provides excellent surgical access. However, this technique involves transection of a large muscle group, and it is believed that this contributes to postoperative pain, reduced inspiratory effort and impairment of arm movement. Attempts have been made to decrease complications by performing muscle-sparing limited thoracotomy using an antero-axillary or anterolateral approach, or, most recently, using video-assisted thoracoscopic surgical (VATS) techniques. Such approaches are associated with significantly less impairment of postoperative vital capacity and improved exercise capacity compared with postero-lateral thoracotomy [69]. In early-stage lung cancer, evidence is accumulating that patients undergoing VATS may experience reduced pain and require shorter hospital stay than those undergoing conventional surgery [70]. However, this is a technically demanding procedure and may require significantly longer surgical time than open surgery; therefore, many surgeons perform limited open thoracotomy, which may be preferable from the standpoint of safety and prognosis. Video-assisted thoracoscopic surgical resection, undertaken by an appropriately skilled surgeon, should be offered to selected patients with clinical stage-one lung cancer (no evidence of tumour spread) where appropriate skill exists, and it is likely that increasing familiarity with the technique and improved training, along with the publication of ongoing research in this field, will lead to increased use of such minimally invasive techniques.

Chest drains impede mobilisation and exacerbate pain in patients after thoracic surgery. There is some evidence from a recently published review that the use of one drain, as opposed to two or more, is associated with reduced postoperative pain scores [71], and when no definite surgical requirement exists the use of one drain alone should be considered. Postoperative air leak is a frequent complication after pulmonary resection for lung cancer. Air leak may prolong the need for chest drainage and delay discharge. Different types of surgical sealants have been developed in an effort to try to prevent or reduce postoperative air leak; however, they are expensive and their efficacy is controversial. A systematic review found that although surgical sealants do reduce postoperative air leak and the time to chest drain removal, this is not always associated with a reduction in length of postoperative hospital stay [72]. Until further research is available, the routine use of sealants cannot be recommended.

Postoperative care

Immediate tracheal extubation favours recovery by allowing early initiation of rehabilitation and re-introduction of oral hydration and nutrition. Furthermore, prolonged mechanical ventilation increases the risk of acute lung injury, pulmonary infection, bronchial stump disruption, bronchopleural fistula and persistent air leakage [73]. Predictors of prolonged postoperative intubation in patients undergoing thoracotomy for lung resection include: intra-operative red blood cell transfusion; high pre-operative serum creatinine level; more extensive surgical resection; and poor pre-operative lung function. Despite this, immediate tracheal extubation should be planned in all patients unless significant complications preclude such a strategy.

Postoperative care in general and pain management and early mobilisation in particular are vital components of any enhanced recovery programme (Table 3). Acute pain after thoracotomy prevents mobilisation and causes patients distress, and is associated with an increased incidence of cardiopulmonary complications, including atelectasis, pneumonia, atrial fibrillation and myocardial ischaemia. Persistent pain is a frequent cause of delayed discharge and is associated with the development of chronic pain syndromes. Epidural analgesia is a core component of many enhanced recovery programmes as it provides excellent pain relief, attenuates the stress response to surgery and accelerates restoration of pulmonary and gastro-intestinal function. Furthermore, epidural analgesia avoids the use of systemic opiods and their associated side-effects. Thoracic epidurals are widely used in patients undergoing thoracotomy and are regarded by many as the ‘gold standard’ for pain relief following pulmonary resection. However, their use is associated with a number of complications including hypotension, urinary retention and motor weakness; these can lead to inappropriate fluid loading, urinary catheterisation and delayed mobilisation, all contrary to the ethos of an enhanced recovery programme. This can be overcome in patients undergoing thoracic surgery by using paravertebral regional blockade, which produces only unilateral sympathetic blockade. A recent meta-analysis [74] has confirmed that paravertebral block can produce comparable analgesia to an epidural, but with a lower incidence of side effects. Adjuncts including ketamine have also proven to be beneficial [75]. It may be preferable to use paravertebral blockade, in association with multimodal analgesia as first-line analgesia, and reserve epidural analgesia for high-risk patients not expected to follow an enhanced recovery programme. These may include patients undergoing chest wall resection or pneumonectomy. There is still controversy over the best method for insertion of the paravertebral catheter, with conflicting evidence regarding ultrasound-guided percutaneous injection vs surgical insertion; further studies in this area are awaited.

Table 3. Recommendations for postoperative care as part of an enhanced recovery programme after thoracic surgery
Area of practiceParameterAction
  1. DVT, deep vein thrombosis

Patient careDVT prophylaxisPharmacological agents should be used in patients who have a low risk of major bleeding [45]
 Drain suctionSuction is not routinely required and may prolong hospital stay [84]
 Drain removalA low threshold for drain removal should be used [83]
RehabilitationMobilisationMobilise as soon as possible after surgery [76]
 PhysiotherapyShould be used to reduce complications and length of stay [48, 78]

Bed-rest is associated with deleterious consequences including impaired pulmonary function, reduction in muscle mass and increased risk of VTE. It is possible to mobilise patients safely as early as 4 h after lung resection surgery, with a positive benefit on the requirement for supplemental oxygen and psychological recovery [76]. Incentive spirometry causes generation of a large and sustained increase in transpulmonary pressure, with consequent expansion of collapsed alveolar units, and has been used widely to prevent or treat postoperative pulmonary complications. However, a systematic review of incentive spirometry in patients undergoing major surgical procedures found no evidence to support its use and it cannot be recommended at present [77]. In contrast, implementation of intensive chest physiotherapy in the peri-operative period has been shown in to reduce the risk of pulmonary complications and resultant hospital length of stay and costs of care after major lung resection [78]. However, the only randomised trial to investigate the impact of physiotherapy after pulmonary resection concluded that there was no benefit over standard care [79]. Non-invasive ventilation has been recommended for the prophylaxis of postoperative respiratory failure in patients who have undergone major surgery or who are at high risk of pulmonary complications. It has been associated with improvement in gas exchange [80] and reduced hospital stay [81], but may increase complications [82].

Early removal of chest drains reduces pain, facilities mobilisation and accelerates recovery following thoracotomy [83]. Drain management is not standardised and there are wide variations in practice, for example whether to use a simple under-water seal or to apply active suction. Some surgeons believe that suction favours the apposition of parietal and visceral pleura, thus promoting the sealing of air leaks, whereas others believe that it increases the volume of air leaks and hinders healing. A recent meta-analysis of six RCTs concluded that it is not necessary to use suction in the absence of a clinically important postoperative space, and that earlier drain removal could result in shorter hospital stays [84]. The majority of thoracic surgeons leave the chest drain in place until fluid output is less than 250 ml per day and the air leak has resolved. Hospital length of stay is often prolonged because drainage is too high or the air leak persists despite the patient being otherwise ready for discharge. A number of studies have demonstrated that a low threshold for drain removal can be used safely, and that Heimlich valves or portable drainage systems can facilitate discharge in the presence of persistent air leak or effusion.


The main component of hospital treatment costs associated with lung cancer in the UK relates to inpatient stay, representing between 62% and 84% [85]. There are, however, wide variations in length of stay following pulmonary resection across the UK (median 5–13 days) [3]. This would suggest that a more uniform approach to the peri-operative care of this group of patients may yield benefits in terms of resource utilisation and efficiency savings, as well as enhancing quality. Protocols that ensure consistent patient management lead to a reduction in complications and consequent delayed hospital discharge, and also improve patient satisfaction.

Despite being pioneered over a decade ago, demonstrated benefit in numerous meta-analyses and support from the Royal College of Surgeons and the Department of Health, the uptake of enhanced recovery programme into surgical and anaesthetic practice has been slow [86]. However, there is no doubt that pressure will continue to be exerted to reduce hospital stay and costs [87], and that, for the moment at least, enhanced recovery programmes are here to stay. Patient selection and safety are paramount, and it is likely that a number of thoracic surgery patients will be considered to be too high-risk to go through such a protocol. Patients undergoing certain procedures such as pneumonectomy or chest wall resection should not be included in such a protocol. A graded implementation process may be considered, starting with relatively low-risk procedures and patients (ASA physical status 1–2), progressing to selected higher-risk patients and expanding to include lobectomy and other longer procedures. Multidisciplinary involvement is vital and surgical, nursing and physiotherapy input will be required. This review of the literature may be used to form the basis for developing a local enhanced recovery programme in thoracic anaesthesia and surgery that may potentially lead to reduced hospital stay and costs in this cohort of patients.

Competing interests

No external funding or competing interests declared. AAK is an Editor of Anaesthesia and this article has undergone an additional external review as a result.