Although COPD is an independent risk factor for perioperative complications in major surgery, somewhat surprisingly specific lung function test values in COPD alone have not been a good predictor of postoperative complications. However, certain closely related factors are important, such as low preoperative SpO₂ or recent pulmonary infection as seen in ARISCAT [2], especially in prolonged abdominal surgery and even more so in thoracic surgery. One study suggests perioperative risk may be increased by COPD with a preoperative FEV1 <70 % predicted or FEV1/FVC ratio <65 % [3]. Diffusing capacity for carbon monoxide is also a guide to severity of emphysematous change in the lung in moderate and severe COPD. Some recent studies have also confirmed COPD as an independent risk factor. In a recent analysis of over 300,000 patients (including 1200 with COPD) in the National Surgical Quality Improvement Program database undergoing abdominal surgery, COPD was independently associated with increased postoperative morbidity and with increased length of stay and mortality in some types of surgery [4]. Postoperative respiratory failure is more common in COPD [5], as is postoperative pneumonia [6]. Surgery during an exacerbation of COPD poses high risks and should only be entertained in an emergency situation. Patients should be optimally treated, until they settle back to baseline lung function. This treatment would normally involve steroids, which – if prolonged – will necessitate additional steroids to reduce the risk of adrenal suppression when surgery is eventually undertaken.

Pulmonary risk during lung resection is associated with abnormal preoperative lung function tests including a preoperative FEV1 under 60 % of predicted in lung resection surgery. DLCO is also useful in predicting risk [7, 8]. Low-risk groups include those with a preoperative FEV1 and DLCO of over 80 % predicted. Predicting postoperative pulmonary function is also important and has involved using formulas such as below that examine the amount of the lung to be resected and its functional contribution.

Predicted postoperative FEV1 (PPOFEV1) = Preoperative FEV1 × (1-(number functional lung segments being removed/total number of functional lung segments, which may be 19)).

One study has suggested that a predicted postoperative FEV1 value and DLCO of >40 % of the predicted normal preoperative values were not associated with mortality. Guidelines from the European Respiratory Society and European Society of Thoracic Surgery use a predicted postoperative FEV1 or DLCO of >30 % as the cutoff, below which death is a likely outcome [9]. Between 60 and 30 % may suggest that exercise testing such as a stair climb or shuttle test is required to better delineate the risk. More precise assessment would involve cardiopulmonary exercise testing. VO₂ max <15 ml/kg produces a high risk of complications [10]. If the predicted postoperative (PPO) VO₂ is likely to be <10 ml/kg, then resection is likely to result in death. Traditional tests such as stair climbing are effective, but unfortunately stairs are not standardized; however, being able to climb less than 12 m increased complications including death [11].

It is important to reduce the postoperative risks in COPD patients by whatever means have been shown to be effective, as postoperative complications are common in patients undergoing one-lung anesthesia especially for lung resection surgery [12–17]. Unfortunately most work in this area has been conducted on abdominal and cardiac surgery and generally in patients without COPD. However some evidence is available, especially prior to lung volume reduction surgery, lung transplant, and, to a lesser extent, prior to lung resection for lung cancer. Preoperative pulmonary rehabilitation improves exercise capacity and dyspnea. One study has looked at the impact on moderate to severe COPD undergoing lung resection due to cancer. Prolonged rehabilitation lasting 1 month proved very difficult to deliver, whereas a ten-session program (including lower extremity endurance training for 20 mins, upper extremity endurance training, strengthening exercises, inspiratory muscle training, and slow breathing) was much more practical and appeared to reduce length of stay and shortened chest drain duration [18].

Despite the inherent potential delay to surgery, several studies have managed to look at pulmonary rehabilitation for 4 weeks prior to lung cancer resection. They have shown a reduction in postoperative respiratory morbidity, including a study by Cesario [19] and an observational study by Bobbio [20]. Unfortunately in one study of lung resection, the patients in the pulmonary rehabilitation arm had a better baseline maximal inspiratory and expiratory pressure [21], which could have influence their future outcomes.

The ability of COPD patients to exercise may not just be related to lung and cardiovascular performance. COPD patients may be unable to exercise because of limb weakness that has occurred secondary to a lack of mobility caused by the lung disease [22, 23]. Some exercise programs can overcome this and produce an improvement in walking distance, dyspnea, pulmonary functional status scale [24], cardiovascular fitness, and leg muscle strength. This then benefits patients in the postoperative period.

COPD is most commonly caused by smoking. Therefore many patients with COPD undergoing surgery are still smokers, which is likely to increase postoperative complications [25]. Smoking cessation 8 weeks prior to surgery appears to be beneficial, improving pulmonary outcomes and wound healing [26, 27]. Frequently, patients will require surgery in less than 8 weeks. It is often thought that stopping smoking close to surgery causes a worsening of pulmonary postoperative complications; however, the evidence for this is not strong [28]. Smokers undergoing surgery will need to stop in the immediate postoperative period, because of the safety requirements of supplemental oxygen. So any concern that stopping may produce respiratory secretions and related issues should not stop a patient from trying to cut down or quit prior to surgery. Consideration of the need for nicotine patches may be important in the postoperative period.

Operations on the thorax or upper abdomen are likely to significantly reduce functional residual capacity. How can this problem be reduced? If a patient can lie flat or in a suitable position for surgery, then a peripheral nerve block (such as in ophthalmic surgery) and regional or spinal/epidural analgesia/anesthesia are likely to have a lesser impact on the lung. General anesthesia using techniques that avoid intubation will reduce related bronchospasm. Unfortunately this will not be suitable for most major surgery. Laparoscopic techniques reduce postoperative pain, which is also helpful in COPD. However insufflation of CO₂ and the compressing effect of gas passed into the abdomen or thorax will add a load onto the respiratory system and contribute to atelectasis.

Induction agents including propofol obtund laryngeal and tracheal reflexes and ketamine has bronchodilator properties. During maintenance, sevoflurane is a good bronchodilator. Desflurane has rapid emergence, which may be helpful. Opioids need to be used carefully to avoid respiratory depression, especially in patients who are already CO₂ retainers. The combination of epidural anesthesia and general anesthesia reduces the incidence of postoperative pneumonia from 16 to 11 % and mortality from 9 to 5 % [29].

Air trapping is a potential major issue in one- and two-lung anesthesia, leading to dynamic hyperinflation, which produces raised intrathoracic pressure in the intraoperative period, with potential respiratory and cardiovascular consequences during and after surgery. This can be visualized as failure to complete exhalation on the monitored flow time graph. Increased expiratory time can be provided by decreasing inspiratory time, increasing I/E ratio, and slowing respiratory rate. Air trapping can become severe, producing an elevation of intrathoracic pressure, which may even require temporary disconnection from the breathing circuit to allow the trapped air to escape. Application of some PEEP may hold airways open during expiration if set to levels equivalent to intrinsic PEEP. However pressures and the tendency to air trapping may vary in different parts of the lung, so the applied PEEP may not suit all lung units and a relatively low PEEP may be more effective in patients with bronchospasm. High FiO₂ leads to absorption atelectasis [30], which appears rapidly after induction of anesthesia [31, 32], becoming exponentially more of an issue when we consider time to increase in atelectasis, especially at end-tidal oxygen levels 90 % and above [33]. This is made even worse in patients with poor V/Q mismatch [34]. However, in the postoperative period, the importance of limiting inspired oxygen concentrations to avoid atelectasis is in some doubt [35], although some have postulated this confusion in the evidence is because of the use of high FiO₂ levels earlier in the anesthesia [34]. However, there are times in COPD when administration of a high FiO₂ may cause problems in patients with a raised PaCO₂, which may lead them to retain even more PaCO₂. Lung protective ventilation during surgery is also important and is dealt with in a separate chapter.

Duration of surgery is very important with a postoperative pneumonia rate of 8 % in operations lasting under 2 h, rising to 40 % for operations over 4 h in length [36]. Complete reversal of neuromuscular blockade is essential to avoid postoperative pulmonary complications [36, 37, 38, 39].

Patients with postoperative lung injury, especially after thoracic surgery, may have a huge increase in mortality (up to 39 %) at 30 days [40]. During one-lung anesthesia, lungs are subject to deflation, atelectasis, reinflation, and the effects of released inflammatory mediators, as well as the threat to anastomoses in terms of airway pressure, vascular supply, and healing.

Thoracic, abdominal, and aortic aneurysm surgeries are frequently complicated by postoperative pulmonary complications. Studies have looked at techniques and exercises that can be introduced or taught to patients in the preoperative period, which can be continued after surgery. Unfortunately, most of these studies have not involved thoracic surgery patients. Despite this, there are some areas where basic principles appear to make a difference. These include mobilization and adequate pain relief. Less invasive surgery can help here. Studies on major abdominal surgery have concluded that postoperative complications including postoperative pneumonia and 30-day mortality are reduced in patients with COPD when thoracic epidural analgesia is used [29].

Early mobilization reduces postoperative pulmonary complications, whereas slow mobilization adds risk with each day of delay [41]. The greatest first day barrier to mobilization was hypotension.

Patients with malnutrition or who have a low albumin are at risk of postoperative complications, and this may require intervention both before and after surgery.

Continuous positive airway pressure (CPAP) has good theoretical advantages after surgery, in that it can increase transpulmonary pressure, improve functional residual capacity, expand collapsed areas, and improve gas exchange. Most studies have looked at major abdominal and cardiac surgeries. Comparisons between very brief periods of CPAP versus 6 h continuous CPAP have shown reduced incidence of pneumonia and re-ventilation with the more sustained treatment [42]. Similarly postoperative CPAP after major abdominal surgery has been shown to reduce atelectasis, pneumonia, and reintubation [43].

Noninvasive ventilation (NIV) could theoretically be useful after thoracic surgery, especially if CO2 retention is present. One concern with both CPAP and NIV is whether there will be an increased incidence of air leaks (air leaks are relatively common after pulmonary lobectomy at around 9.7 %). Fortunately the small number of studies performed on thoracic surgery patients does not appear to have found an increase [44, 45]. NIV has been used after thoracic surgery and has been found to increase arterial oxygenation.